.. :. .

,'

athematics 11 B.Sc. CSIT Second Semester

· Binod·Prasad Dhakal, Ph.lJ.

Surendra Path~k,"-Ph.D.
Ram·',esh Gautam
You·brc}j Gai re
Ram Prasa:.d Sub~di (Mukesh) .
 MATHEMATICS - II
B.Sc. CSIT II Semester

B_inod Prasad Dhaka!, Ph.D .

P rofessor
Central Department of Education (Mathematics),
T.U., Kirtipur

Surendra Raj Pathak, Ph.D. Ramesh Gautam, M.Phil.

Associate Professor Lecturer
Amrit Science Campus Ratna Rajya Lax.mi Campus,
T. U., Kathmandu Kathmandu
Sagarmatha College of Science and
Technology, Lalitpur

Youbraj Gaire, M. Phll. Ram Prasad Subedi (Mukesh)

Faculty Mel!lber Faculty Member
Central Department of Mathematics Central Department of Mathematics
T.U., Kirtipur T.U ., Kirtipur
Nepal En&ineering College,
Bhaktapur

KEC Publication and Distribution (P.) ltd.

;I::~ Kathmandu , Nepal
~,~ Phone: 01-4168301

-
 MATHEMATICS-II
B.Sc. CSIT, II SEMESTER

ISBN No.
978 - 9937-8925-7-5

Publisher
KEC PUBLICATION AND DISTRIBUTION (P.) LTD.
Kathmandu
Ph. 01-4168301
l•
Distributor
KEC PUBLICATION AND DISTRIBUTION (P.) LTD.
Kathmandu
Ph.0l-4168301,01-4241777

kecpublication.com .np

blicationl 4@grnai].com ·

Edition
Second Edition 2076
Third Edition 2022 (Revised)
 Preface

Basicall y, Mathematics II includes two parts of algebra - Linear Algebra and Modern
:'1gebra .. There is wide application of Linear Algebra in business, economics, engineering,
mfo~atton technology, physics, computer science, ecology, sociology, demograph y and
genetics. On the other hand , Modem Algebra is applicable in the field of cryptography which
has wide range of application in the field of computer science. Tribhuvan University designed a
course Mathematics II in B.Sc. CSIT second semester but a suitable textbook was a long felt need .
Although there exists many books of this course by foreign authors but there are no books which
are strictly based on the syllabus of Tribhuvan University. This book is written according to the
syllabus of B.Sc. CSIT second semester so it fulfill the need of students. The main goal of this
book is to help students master the basic concepts and skills they will use latter in their careers.
This is also a useful reference book for students of other faculties such as B.Sc., B.Ed ., B.A. and
B.E. of different universities of Nepal. In this book the subject matter is presented in simple and
sequential order that can be u seful to create interest among the readers. Each chapter covers
definitions, examples and theorems with numerical examples so that readers feel comfortable to
grasp the contents of the book.
We would like to thanks KEC Publication and Distribution family for making efforts to
publish this book. We are grateful to all our friends, teachers, seniors and well wishers and family
members who have supported and encouraged us for this task.
An y sugges tions, feedback fo r improvement will be highly appreciated and gratefully
acknowledged.
[email protected]

Authors
 Mathematics- II Syllabus
Course Tille: Mathematics fl
Course No: MTH 163 Full Marks: 80
Nature of the Course: TI1eor and Practice Pass Mark : 32
Semester: II Credit Hrs: 3
Course Objectives
TI1e course contai ns conce t d I . .
Unear equah· d . . P s an tee inique of lmear algebra. The course topics includ e systems of
va lu e deco ons,·t·elermf mants '. vectors an vec or spaces, e1·gen va lues and eigenvectors,
d t · ·
and s111gu l,H
mpos1 ion o a matrix.
Course Description
TI,e main objective of the c · k f .. . .
al b o urse is to ma e am1hanze with the concepts and techniques of linea r
ge ra, dso Ive sy stem of linea r equation w ith Ga uss-Jordon method to impart knowledge of vector
spacetr· an]" subspace
. ' eigenval
. ue and eigenvectors
· of a matri x and ge't the idea of diagonalization of a
ma 1~, mear programmmg, Group, Ring, and Field.
Course Details
1. Linear Equations in Linear Algebra [5 h ]
Definition of Ii ti C • . . . . . . ours
. ne_a r equa on, ons1stent and mcons1stent; Matnx notahon, Solv111g a lmear system,
Exi stence and umqu eness questions, Related problems (Ex. 1.l(no. 1-22, 29-32)), Definition of echelon
~~~m. and redu~e row echelon _form, ~xamples, Pivot posihons, Related problems (Ex. 1.2 (no. 1-14,
. 20)) ~ectors m ~' G~ometnc description of R2, Vectors 111 R3, Geometric description of R3, Vectors
m Rn, Lmear combmatJons, A geometric description of span v. Ex.1.3 (no. 1-18) The rr:iatrix equation
A_ x = _b, Ex. 1.4 (no. 1-1 5), Application of linear systems, A homogeneous system in economics,
Linear mdependence, Linear independence of matrix columns, Sets of one or two vectors and related
problems (Ex. 1.7 (no. J -20)) .
2. Transformation [4 hours]
Introduction to linear transformations, Matrix transformations, Linear transformations, Related
problems (Ex. 1.8 (no 1- 19))., The matrix of a linear transformation, Geometric linear transformation
of R2, One to one a nd onto mapping, Related problems (Ex. 1.9 (no. 1-10)), Linear models in business,
science, and engineering (example 1, 2, and 3) .
3. Matrix Algebra [5 hours]
Matrix operations, Sum and scalar multiples, Matrix multiplication, Properties of matrix
multiplication, The transpose of a ma tr ix, Related theorems and related problems( Ex. 2.1 (no. 1-12,
16, 17, 27,28)), The inverse of _a matrix, Invertible, Singular and non singular matrix, Elementary
matrices, Characterizati ons of in vertible matrices, Related problems (Ex. 2.2 (no. 1-7)), Partitioned
matrices, Addition and scalar multiplicatio n multiplication of partitioned matrices, Multiplication of
partitioned matrices, I1iverses of p artitioned matrices, Related problems (Ex. 2.3 (no. 1-10), Ex. 2.4 (no.
1-10)), Matrix fa ctori za tions, The LU fac torization, Related problems (Ex. 2.5 (no. 1-15)), The Leontief
input- outp ut mod e l (exampl e 1 and 2), Subspaces of Rn, Column space and null space of a matrix,
Related problems (Ex. 2.8 (n o. 1-20, 23-26)), Dimension and rank, Coordinate systems, The dimeI;lsion
of subspace, The ra n k Lheorem (no proof), The basis theorem (no proof) and invertible matrix theorem
(no Proof), Ex. 2.9 (no. 1-7)
4. Determinants [4 hours}
Introdu ction to determinants, Properties of determinants, Determinants and matrix products, Related
problems (Ex. 3.1 (no. 1-38) Ex. 3.2 (no. 1-26), Cramer's rule, Volume, and linear transformations, An
inverse formula and Related problems (Ex . 3.3 (no. 1-22)), Determinants as area or volume (example
4) Linear transformation .
5. ~~~~ ~~~
Vector spaces, Subspaces, Zero s ubspaces, A subspace spanne? by a set (Theorem 1 (no proof)), The
null space of a m atrix, Theorem 2, The column space of a ma_tr1x, The contrast between nul A and col
Kernel and ra nge of a linear transformation (example 8, 9), Related problems(Ex.
A (examp1e 5, 6, 7) , . . . (
_ (no. 1-18) (Ex.4. 2 (no. 1-24), Lmearly independent sets; Bases, Examples, The spanrung theorem no
41

-
 proof), Bases for nul A and col A, Theorem 6 (no proof), Coordinate systems, The unique representation
theorem, The coordinate mapping, Related problems (Ex. 4.3 (no. 1-20) Ex. 4.4 (no. 1-14)).
6. Vector Space Continued [4 hours]
The dimension of a vector space, Theorem 9, 10 ( no proof, concept only), Subspace of a finHe
dimensional space, Theorem 11,12 (no proof), The dimension of nul A and col A, Related problems
(Ex. 4.5 (no. 1-17)), The row space, Theorem 13 (no proof), The rank theorem (no proof), The invertible
matrix theorem (No proof), Related problems (Ex. 4.6 (no. 1-7)), Change of basis, Theorem 15 (no
proof), O,ange of basis in Rn, Application to difference equations (example 1, 2, 4), Related problems(
Ex. 4.7 (no. 1-10)), Applications to Markov chains (example 1, 2) .
7. Eigenvalues and Eigenvectors: [5 hours]
Eigenvectors and eigenvalues, Theorem 1, 2, Related problems (Ex. 5.1 (no. 1-20)), The characteristic
equation, Similarity, Application to dynamical systems (example 5), Related problems (Ex. 5.2 (no.
1-14) Ex. 5.3 (no. 1-18)), Diagonalization, The diagonatization theorem (no proof), Diagonalizing
matrices (theorem 6), Eigenvectors and linear transformations, The matrix of a linear transformation,
Linear transformations on Rn, Diagonal matrix·· representation (no Proof), Complex eigenvalues,
Real and imaginary parts of vectors, Related problems (Ex. 5.4 (no. 1-7) ((Ex. 5.5 (no. 1-20)), Discrete
dynamical systems (example 1,2, 3, 4), Applications to differential equations (example 1, 2) .
8. Orthogonality and Least Squares [5 hours]
The inner product,theorem 1, The length of a vector, distance, Orthogonal vectors, Orthogonal
complements, The pythagorean theorem, Theorem 3 (no Proof), Angles in R2 and R3, Related
· problems( Ex. 6.1 (no. 1-18), Orthqgonal sets, Theorem 4, Orthogonal basis, Theorem 5, An orthogonal
projection, Orthonormal sets, Theorem 6, Theorem 7(no proof), Related problems( Ex. 6.2 ((no. 1-22)),
Theorem 8 (no proof)·, The Gram- Schmidt process, Theorem 11 (no proof), Least square problems,
Solution of the general least square problem, Theorem 13 (no proof), Theorem 14 (no proof), Theorem
15 (no proof), Related problems (Ex. 6.3 (no. 1-18) Ex. 6.4 (no. 1-8)), Application to linear models
(example 1, 2, 3),
Inner product spaces, Length, Distance, and orthogonality; Cauchy -Schwarz inequality, Related
problems (Ex. 6.5 (no. 1-14) (Ex. 6.7 (no: 1-6)), Application of inner produ~t spaces (example 1, 2) .

9. Groups and Subgroups [5 hours]

Sets, Introduction and examples, Complex numbers, Multiplication of complex numbers, Related
problems, Binary operations, Definitions and examples, Com.mutative, Associative, Related problems,
Isomorphic binary structures, Definition and examples, Uniqueness of identity element (no proof),
Exercises 3 (1 - 10), Definition of group and examples, Abelian group, Elementary properties of
groups, Finite groups and group table, Exercises 4 (1- 6), Order of a group, Subgroups and examples,
Proper and improper subgroup, Trivial and·non trivial subgroup, Examples, Theorem 5.14 (no proof),
Generator of a group, Definition of cyclic group and examples, Theorem 5.17 (no proof), Exercises 5
(1-12, 14-19, 21,26, 27) .
10. Rings and Fields [4 hours]
Definition of ring and basic properties, Examples, Theorem 18.8, Exercises 18 (1-12), Divisors of zero
and examples, Integral domain, Examples, DefinHion of field, Theorem 19.9, Theorem 19.11 (no proof),
Corollary 19.12 (no Proof), Exercises 19 (1- 4) .
Text Books: · ·
1. Linear Algebra and Its Applications, David C. Lay, 4th Edition, Pearson Addison Wesley'.
2. Linear Algebra and Its Applications, Gilbert Strang, 4th Edition, Addison, CENGAGE Learning.
3. A First course in Abstract Algebra, John B. Fraleigh, 7th Edition, Pearson .
• LINEAR EQUATIONS IN LINEAR ALGEBRA

1.1 Systems of Linear Equations .... .... ............ .. .................. ................................. .............. 2
1.2 Row Redu ction and Echelon Forms .......................................................................... 7
□ Exercise 1.1 ..................:...... .. ...... .. .. ... ............ ...................... .. .. ............ .. ....... .... ...... ..... 15
1.3 Vector Equ ations ........................... ............. ... .................. ........ ............... ................. ... 17
1.4 The Matrix equation Ax = b ................................................... :.................................. 21
□ Exercise 1.2 ............... .......... .. ............ ........................................................................... 25
1.5 Solution Sets of Linear Systems ................... .......................... ................................... 27
□ Exercise 1.3 ............... ........................... .. ...................................................................... 30
1.6 Applications of Linear System ............ ...... ................................... .......................... .. 31
1.7 Linear Indepe11dence .. ............................ :.................................................................. 34
0 Exercise 1.4 .... .. ............................................................................................................ 39

2.
enapte.'t
TRANSFORMATIONS

2.1 Introduction to Linear Transformations ................................................................ 42

□ Exercise 2.1 ........... •.. •••••••····························································································•46

2.2 ·The Matrix of a Linear Transformations .. .......... ....... ................. ....... .. .................... 47
□ Exercise 2.2 .................................................................................................................. 55
2.3 Linear Models in Business, Science and Engineering ........................................... 57
 MATRIX ALGEBRA
Introduction ... ... .. ... ... ........ ... .... ............ ...... .. ..... ... ,........... .. ........... ............................... 62
3.1 Matrix Operations .. ... ................. .. .. ......... .. ..... ....... .. ...... .. ... ..... ....... ........... ................. 62
3.2 The Inverse of a Matrix ...... ... ... ...... ... ........ .......... .... .. ..... ............ .. .... ........... ............... 67
3.3 Characterizations of Invertible Matrices .. ... ... .................. ........... ............................ 72
D Exercise 3.1 ..... .. ........................ ........ ........ ........................ ............ ... ....... ..................... 73
3.4 Partitioned Matrices .................. ... ........... ......... ..... ...... ... ........ ... ..... ............. ......... .... .. 75
D Exercise 3.2 ... ....................... ........... ........... ..... .... ............ .... ........... ... ........................ ... 79
3.5 Matrix Factorizations ............... ... ........ ........ .................. ...... .................................... .. . 80
3.6 The Leontief Input Output Model ............. ............................... .. ... ......... ................. 83
3.7 Application s of Matrix Algebra to Computer Graphics ......... .............................. 87
3.8 Subspaces of Rn ...................... .................. ........................ .. ...... .... ..... ..... ............. ...... . 89
3.9 Dimension and Rank ................................................... ..............-................................ 90
□ Exercise 3.3 ................................................ ... ............. ........... ...... .. .. ............................. 91

4
efiapte't
DETERMINANTS

4.1 Introduction to Determinants .. ......... .................................... .................................... 94

□ Exercise 4.1 .. ................................. ••.••••••••••••·...... ••••••.•· •... •.... •...................................... 97
4.2 Properties of Determinants .... ................. ........ .... ... .......... ......... ................................ 99
□ Exercise 4.2 ................... .... .•·· ···•·············· ················· ········ ···· ···················· •·· •········ •····· 104
4.3 Cramer's Rule, Volume and Linear Transformations ......................................... 106
□ Exercise 4.3 ......... ............... ...... ..... ........... .......... ...... ... .............................. :................ 111
♦ VECTOR SPACES

Introduction .......... ... ........ ......... ........ .. .. .. ... ...... ... .... ... .................... .... ........................ 114

5.1 Vector Space and Subspace ................. .. ........ .............................. ..................... ....... 114

□ Exercise 5.1 ··············· ···· ·· ················· ········ ·· ········· ···· ··· ··················· ···· ······· ··········· ···· ··· 119

5.2 Null Space and Colu1nn Space, and Linear Transformation ........ ......... ......... ... 122

□ Exercise 5.2 ·· ········ ·········· ························································· ·· ·· ······ ······················ ··· 128

5.3 Linearly Independent Sets, Bases ....... .......... .......................................................... 130

D Exercise 5.3 ....................................................... :........................................................ 131

5.4 Coordinate System .............. ...................................................................... ..... ... ....... 135

□ Exercise 5.4 ..... ......... ... .. ............. .................... .. ......... .. ................................. .............. 139

6
e/iapte't- VECTOR SPACES CONTINUED

Introductio11 ........... ............................ ..................... ..... .. ............................................ 142

6.1 The Dimension of a Vector Sp ace .... ...... ... ......................................... .......... .. ........ 142
0 Exercise 6.1 .............................. .... .. ....... ........................................................ ............. 144
6.2 Ran.k ...... .......................................... ............................................................................ 145

D Exercise 6.2 ...... .............. ............ :........................................ ........................................ 146

6.3 Change of Basis ................. ......... .. ............................................................................. 148

6.4 Applications to Difference Equ ations ........... .............. ............... ............... ............. 150

6.5 Application of Markov chains ........ .............. ............................................. ............. 151

□ Exercise 6.3 ............ ........ ....... •· ••·· •··•·· ............................................. ............................ 153
♦ EIGENVALUES AND EIGENVECTORS --.....

Introduction ...... ........................... .......................................... .............. .... .......... ....... ... .... ......... lS6
- .......... ... .............. ................. 156
7.1 Eigen Value and Eigen Vectors ......... .. ........ .... ...... ... .. ........ .. ..... .

□ Exercise 7.1 ........... ............. .... .... ............ ......... ....... .... ...... ..... ... .H•• ······ ····················· ·· 160

7.2 · ti.c Equ ati·on ......... ... .... .............. ••. ••· •· •················· ····· ·· · ·· · ····························· ··· 162
The Charac teris
□ Exercl·se 7.2 ·························································.......... .. ........ .. ............... ..... ......... ... . 167
7.3 Diagonalization ....... ....................... ........... .._........... ..................................... .. ..... ...................... 168

□ Exercise 7.3 ........... ... .................... .. ........ ........... ........ ..... ............ ........ ;....... ... .... ......... 178

7.4 Eigenvectors and Linear Transformation .. ...... .. ..... ... ......... ...................... ............ ...... .......... 179

□ ••,. ••• ••• ••• ••• ••••. ••.•••• • ••••.•.••••••• •,•· · · ·.: · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ···· 183
Exercise-7.4 ...........,.. .........! ••• : .;r• • • .

7.5 Complex Eigenvalues .............. .. :......:........... ........ ........... ...............:.. ................... ... .............-:- 185

D Exercise 7.5 ........................................ ... ................. .......... ...... ........... .. ..... ................. . 189

7.6 Discrete Dynamical Systems ............. ....... ...... ....... ...................... ... ......... .. .... .... ......... ............ 190

D Exercise 7.6 ............ :.... .... .......... · 196

·················································································

ORTHOGONALITY AND LEAST SQUARES

8.1
Inner Product, Length and O th .
. r ogonahty ............................. ..........................:.... 196
□ ) Exercise 81
. ...................... ......... .. .. ...... .. 202
8.2 Orthogonal Sets ... ..................... ......... ..... ... ....... ......... ........... .
Exercise 8.2 .. ............................................... .. ............ .............................. 203
□
8.3 Orthogonal~~~~~~~~-~·~~· -· - ·················· ··· ··· ···· ······· ············ ····· ·· ··· ········· ·· ····· ···· ··· 209
············
············································· ······························ ····· 211
 0 Exercise 8.7 .... .... ..... ...... ..... ... ............... .... ... ......... ... .... ...... .... .... .............. .... .... ... ... .. .. . 218
8.4 The Gram Schmidt Process ...... .......... ................. .... ........ .......... .... ........ ....... ..... ... ... 220
□ Exercise 8.4 224
····· ······· ······· ············ ··· ····· ············ ··············· ·· ·················· ·········· ·········· ·· ····
8.5
Least Squares Problems ....... .. ............. ............... ..... ... .. ... ..... .... .......... ....... .... ....... ... . 225
□ Exercise 8.7 229
·· ·· ················ ········ ······· ························ ········································ ··· ···· ······
8.6
Applications to Linear Models .................. ........ ............... .. ... ....... ......................... . 231
□ Exercise 8.7 ....... ........................................ ........... .. ....... ................... ... .. .... ... .. .... ........ 234
8.7
In11er Product Space .... ... .. ...... ... ....... ... .... ... .......... ... ... ...... ....... ..... :.... .... ........ .. ..... ..... 235
□ Exercise 8.7 ... ... ..... .. .... .............. .. ..... ............ .. ...................... ......... ....... ........ ....... ..... .. 243

GROUP AND SUBGROUPS

9.1. Sets and Set Operati ons ......... ...... ...................... .. ...................... .. .. .. ....... ............ ......... 246
9.2 Mappir1gs ........... .......... .......... ........... ................. ........... ............ .. ... ..... ... ................. ........ 248
9.3 Group .. .... .............. ........................... ......... .. ...................................... ..... ..... ... ........ ..... ...... 250
□ E xerc1se 1 ............................ .. .................. ........ ............................ ........... .......... .... 262
· 9 ....
9.4 Subgroups ..... ......... ..... ..... ... .......................... .............····················· ·· ········ ············ ···· ···· ·· 263
□ . 9.2 ..............._. .................. .... .. ...... .. ········· ······· ·· ········ ···· ..... ........................ ... .. 265
Exercise

10
eliapte.i
RING AND FIELD

.
10.1 Ring and Basic Properties.... .............. ....... ..... ..... ....................................... ...... .. ......... ........... ... ... 268
274
10.2 Properties of Rin g ...................................... .......................................... ............................. ......... ...
..................... ...... ........ ........ 275
10.3 Field ........................ .............................................. ······························· 179
□ Exercise - 10 ..... :...... ............. ..... ................. ............................... .................................. ..... ......... 281

□ Bibliography...................................... 282

□ Model Question ................................. 284

TU Examination ..................................................... .·············································
□
 I ' I .

LINEAR EQUATIONS IN ' - . . . . . .

LINEAR ALGEBRA

LEARNING OUTCOMES
After studying this chapter you should be able to:
~ Systems of Linear Equations
~ Row Reduction and Echelon Forms
~ Vector Equations
~ The Matrix equation Ax = b
a. Applications of Linear System
a. Linear Independence
 MATHEMATICS-D

I.I SYSTEMS OF I..INEAII EQUATIONS ~

. b d it helps to introduce some of central
System of linear eqµations lie at the heart m alge ra_ an Also the system of linear equations is
concepts of linear algebra in simple ~d con~ete setting, , .
equivalent to vector equation and matnx equation. --------=--~~~----;;----:rr:-;~~
on inear Equatloi\ with Real or Co'!'pl~x Co,tflcient)
;JI: •) th i
iPJ:111" onuation ln the variables Xt, X2, • ~ :,,
i • .

· ~ -~an.equ~tio.n ip. e orm

-~ ... , ' 1 {. '
,, ~...- + " ~>½A,~J
•4
~ ~r ~ •, :t
a1X1 + ~x2 + ... + anXn =b'
~ tiel'll!Jc&S.- --:~-=...ts a a ..· a.. arid .the vai~e,: b are re~· or "complex nun\bers
'd UR:'~• t, 2, • • •f n < ,.., t ,..,,, : 'I c /"'*
known in advance and~ may be any positive-integer. •1/ :

Example 1: The equations,

4x1 -Sx2 = 7 ➔ Linear
2(V6 - Xt) - X2 + X3 =Q ➔ Linear

'/xi- X2 =6 ➔ Not linear

Example 2: The systems,

(a) x1 + Sx2 = 7 4x3 = 8 (b) X2 -

2x1 + 7x2 = 5. 2x1 - 3x2 ·+ 2x3 = 1

5x1 - Bx2 + 7x3 = 1.
are examples of system of linear equations.

Di.fin . (Solution Jl),e ,.-m of Linear !q~ons) •t-:'.""':~~-~ - - ; - - ~ ~ : - - ~ - - - -

: ~ ol the system is a list of values (x1, x~ ... , ~•) ~£ ~~~ .~ ~ .tisfies the giv
... ' ...
For an example, the linear system
x1 + Sx2 = t

2x1 + 7X3 = 5
is satisfied by a list (x1, x2) = (-8, 3). So, (-8, 3) is the solution of the syst em.

Note that the solution of a linear system does not necessarilY exISt
. This
may not have solution. And, sometimes the syst ha · means, some linear system
em s exactly O l •
:S}'Stem has infinitely many solutions. ne so ution and sometimes the
 Linear l:quatlona In Linear Alpbra w ICaAmR 11
Definition (System of Homogeneo U

. ... (ii)

is system of
N"ote that if any bi "' 0 for i _ :1 2 .·
equations. · __ -. · _- _ <,. ,': .
th. . ~ . .
, m en the system (i) is called non hom<?geneous linear
, .l~. •

1. Remember that the h omogeneous linear . .

equations .always has solution
(xi, x2, · · ., Xn) = (0, 0, •.. , 0) which we know as trivial solution or zero solution.
2. The solution of system (ii) is the particular·solution of system (i). ,, .
Definition -~onsistent an
.. ....
.

A ~ . . of linear, eq-q~_ · ·
ir,m"\it.ej~·ll@lly solutio
Graphical Representation of Consistency of Linear Equations · ·
I

Geometrically, linear equation represents a straight line. A system ~f linear equations is set of straight
lines in where the number of lines is exactly equal to number of equations _in•the system. The ·system of
linear equations is consistent if the equations have a common (at least one) solution 'This means, if the
lines of the system intersect each other at a point (at least one point) then the system is consistent and if
they all are not intersected at common point then the system is inconsistent. For instance,
. . . . _,. . . . . . . .. n.., ~
' (i)' Toe·equation of lines 11, hand h in' figure (a) is consistent and has fX~~tly;on'e solution oecause
. , 1 .

the lines are intersect at a point.

. ,'. •
(ii) The equations of lines 11 and h are parallel. So, the system of equations of h and l-i, i~ inconsistent,
in figure (b).
(iii) Th; line ft is overl~p to 12• This means the line·s ·have infinit~ly many_ point~ of inte~s~t. So, the
. ns of line 11 and Ii has infinitely many solutions and 1s consistence, m figure (c).
system of equa tio . .
y ,, . . y . -:. ,1,
y

'2
h (c)
(b)
(a)
 11ATJIBIIAflC8-II r.i

Example 3: Consider a system

X1-2x2 • -1

X1-3X2 =-3
Lines satisfy by the point (3, 2). So, the
system is consistent and has exactly one
solution because the lines intersect each
other only at a point.
Example 4: Consider a system
X2
X1 + 2x2 = -1
X1 + 2x2 • 2
This shows that lines represent by the
· system, are parallel. So, the system has no
solution. The graphical representation of
the system shows the lines are parallel. So,
the system is inconsistent.

Example 5: Consider a system · X2

X1-X2 =1
-3x1 + 3x2 = -3
This system has the overlapping lines. So,
the system has infinite solutions and is
consistence.

Example 6: Why the system xi = 8 is inconsistent? Giv~ . graphical

' ' t
representation. · ' . 1

X2
Solution:
.)

Given system is,

... (i)
-3x1 + 9x2 = 8 · ... (ii)
Oearly the Jin~ (i) passes through the points
(4, 0) and (0, -4/3) and the line (ii) passes
through the ppints (-8/3, 0) and (0, 8/9).
Then graph of the system is,

Here, the lines ,are parallel. So, the system has no

solution. Therefore, the system is inconsistent.
 Unear Bquatlou ID Unear Algebra W eijnm M·
If the matrix notation involves only the coefficients of variables then the matrix is called
coefficient matrix. And, if the matrix notation involves the coefficients of linear system as well
as constant value then the matrix is called augmented matrix of the system.
Consider a system of linear equations,
aux1 + a12x2 + ... + a1nXn = bi }
a21x1 + a22x2 + ... + a2nxn = bi

am1X1 + am2x2 + ... + amnXn = bm

can be written in matrix equation as
Ax=b
where A, x and b are matrix represent the above system,

[au
JI:~] [fil
a12
a21 a22
~t1
11
amt am2 =
Note that

A=
[au a21
a12
a22 a]
a2n
➔ Coefficient matrix of (i)
amt am2 a =
/ Ii
.,
• J

--+xi
x= '
[::] ➔ Variable or output matruc of (i)

,hicaJ
b=
[fil '. ➔ Resultant 0~ input matrix of (i) I

au tl
an
[A: b] = ...
[
am1 am2 .. ,amn :
.~J ,I

➔ Aug,riented lllitrix-Of (i)

1

Note: In matrix equation if b = 0 then the equation reduces to ~ = 0 which is a system of

homogeneous linear equations. ' · ' ·1'
Example 7: Consider a linear system
x2 +4%3=-S
%1 + 3%2 + 5%3 = -2
3x1 +7x2+7x3=6
Solution: .. .
Then, the matrix notation with coefficients each variables in the form,

1
o 1 4]
3 5
·
[3 7 7
, II ..
is the coefficient matrix.
 MATHEMATICS-II
,Ji I nit · • 'f ·,rh
·And, the matrix notation of the system
:Ti,
I

I. '[~
3
!:
7 7
=;J
6
is the augmented matrix.
Elementary Row Operations t
' < '

·
1. (Replacement) Replace one row by the sum •
of ~tself and a mu1ti'P· Ie ·O· f another
. ·
row. 2

2. (Interchange) Interchange two rows. : ' . · ,

~ ' ~- ~ ';, < /"
3. (Scaling) Multiply all entries·in ~a ro~ by a nonzeto const~nt., •

Example 8: Solve the systems of linear equation by ~lementary ~ow operation.

X1-3X3=8

2x1 + 2X2 + 9X3 = 7

X2 + 5X3 = -2.
Solution:
Given system is,
X1 ...! Jx3 :i:: 8' '
2x1 + 2x2 + 9x3 = 7
x2 + Sx3 = -2.
The matrix notatio~ ~f tlJ.e _sy;stem is, , 1 ,.

[ ~ ; ~]
~

~•[r"f1
0 1 5 -2

Apply R2 ➔ R, ,-2R1 ~,r]educes to

0 1 5 -2

Apply R.➔~ ; & ~[r f I'~bixI]d:::laI

th
form.
0 0 -6 , .. 5 ,,
The equation form of the matrix notation is,
X1 - 3X3=8 , .. (i)
2x2 + 15x3 = -9 ... (ii)
-Sx3 = 5 - ... (m)
From (iii), we get X3 = -1. ' .
Substituting the value X3 = -1 in (ii) then it gives,
2x2-15 = -9.
⇒ x2=3. ..r ' . .,
I

And, substituting the value X3 = -1 in (i) then it gives,

Xt +3=8.
⇒ Xt = 5.
Thus, the solution of the given linear system is (x1, x2, x3) = (5, 3, -1).
 Uneu Bqaatlona In Linear Alpbra .W iCeAm,R it
J.2 ROW REDIJCTION AND EOIELON FORMS

Row Echelon Form)

ft!C~ p]~ ,r;naitrix is in echelon form (or row echelon form) if

nonzero rows are above any rows of all zeros.

ch leading entry of a row is in a column to the righ
:ve it.

entries in a colutnrt below a lead_ing entry are zeros.

matrix ill echelon form satisfies the following ·
uced echelon form (or reduced row ech · '·
eleading entry in each nonzero.row.,is 1
5. Each leading 1 is the only nonzero entry · ·
The following matrices are in echelon form:
2 -3 2 1 0
0 1 -4 18 ] , 0 1
[ [
0 0 0 5/2 0 0
-
In which the second matrix is in reduced echelon form .
.. Ee
tri'>C
The following matrices are in echelon form. The leading entries (■) -may have any nonzero value; the
starred entries(*) may have any value (including zero).

~(;' '
0
0 0

The following matrices are in reduced echelon form because the leading entries are 1' s, and there are
O's below and above each leading 1.

,,,..
'..,, ~1'"·0 ·o :To.··-- .- '..•_~, ..
. () •, 1
00 ·~·0 o o· • *
I

0 0 0 0 · 1. 0 .. • ·
o o ,o
[o 0 0 Q, ' 0 O ~ 1 , ' ~ ''W!fl ~i
l,,
· a· 0 0 0 0 . 0 0 0...
ifTf.tt·'-
t;;i-"1. ~~~~
.

-
' 1.,. ,

(1) Different sequences (operations or formulae) of row operation gives _different echelon
matrix of any non-zero ma~. This means, a matrix may have more than one echelon 1

matrix as the operated row operations.

(2) Remember that a non-zero matrix have a unique (i.e. one and only one or exactly one)
reduced echelon matrix.
 MATBBIIATIC&-11

This note leads the concept of following theorem.

_,- I

A= -2 -3 0 3 -1
1 4 5 -9 -7
[
d t the pivot columns of A.
Here row reduces the matrix A below to 'echelon form, an 1oca e

4 9
3 1
0 -3 -6
-1 -2 -1
l
. 'ti A nonzero entry, or pivot, must be
The top of the leftmost nonzero column is the first pivot posi on. d (because the mental
placed in this position. A good choice is to interchange rows 1 an 4
computations in the next step~ not involve fractions). 1
Pivot
1
1.J 4 5 -9 -7
-1 -2 -1 3 1
-2 -3 o 3 -r I /,

., 0 -3 -6 4 9
''
L '
Pivot c'olu~ n
.
Create zeros below the pivot, 1, by adding multiples of the first row to the ~ows below, and obtain
matrix (1) below. The pivot position in the second row must be as far left as possible-namely, in the
second column. Choose the 2 'in this position as _the next pivot. ,.
,,,
4
FPivot
S -9 1-:

[i -7]
.>_\/' ' 7. l . I

2 4 -6 -6 ' • • •~ J. • ••

5 10 -15 -15
. -3 -6 4 9 ... (1)

L Next pivot colu~

Add -5/2 times row 2 to row 3, and add 3/2 times row 2 to row 4 .
..
1 4 5 -9 ;_7
0 2 4 -6 -6 '· I '~ I

... (2)
0 0 0 0 0
0 0 0 -5 0 ~,
There is no way to create a leading entry in column 31 C/Ve can't use row 1 or 2 because doing so
would destroy the echelon arrangement of the leading entries already produced.) However, if we
interchange rows 3 and 4, we can produce a leading entry in column 4.

,5 .-9}Pivot
1 4 -7 •0 *
• **
0 2 4 -6 -6
* *
0 0 0 -5 0
General form : * *
0 0 0 0 0
0
0
0 0
0 0
•0 *
0
t t f Pivot columns
 Linear Equations In Uneu Alpbnl w e- ii
The matrix is in echelon form and thus reveals that col~ 1, 2, and 4 of A are pivot columns.

Pivot position.
- 3 -6 4
[ - 0I -2 -1
A= ~ -3
4
0
3
3
5 -9 -7
_:] ... (3)

Pivot columns
A pivot, as in above, is a nonzero number in a pivot position that is used as needed to create zeros
via row operations. The pivots as above were 1, 2, and -5. Notice that these numbers are not the
same as the actual elements of A in the highlighted pivot positions shown in (3).
~ ~ -(Pivot Posfffon
A pivot position in a matmafA is a- location in .A that ~orr~ds td a l~~g 1, in.the reduced!
echelon farm of A. A ivot colt.ttnn i$ a ~olumn Qf ~ t COQ~ a pj.yot · ·~

1leduciion Alg
, tithm hel
,, ,. ¥

I Example 9: (i) Change into reduced echelon form of the matrvc

,,.,.

I ~ _37
[ 3 -9
t]
12
(ii) Reduce the augmented matrix of the system in reduced echelon form
I

[3 _i ~
! ! ~]
O O 7 -5

Solution:
(i) Given matrix is

~6]
12
 MATHEMATICS-D

Interchanging Rr and ~ to make non-zero pivot then, · ,

3 -7
- 0 3
[
3 -9

Apply R2 ➔ R2/3, ~➔ ~ - R1 then the above matrix reduces to

-7

-rn 1
0 f]
Apply R1 ➔R1 + 7R2 then the above matrix reduces to
0

-rn 1
0 f]
1his is a reduced echelon form of given matrix.
(ii) Exc
Interchanging R1 and R2 to make non-zero pivot then,
0, 3 0

-[i 1
-2
0
Q,
3
0
-3
2
7 IJ
So

Apply ~ ➔ ~ -3R1 then the above matrix reduces to

r~
~ 0 -2
0
0
1

0
3
0
3
-9
Q.
-3
2
7.
2]
3
l
-11
.

Apply R3 ➔ R3 + 2R2 then the above ma.t rix reduces to

~ .f. ~ -~ ; ]
~ [g g. ! 1 _rt

Apply~ ➔ & + 3RJ then the above matrix reduces to

Jil~ r. i ~ o o -5
i]isaechelonform
10 . .

Now, for reduced echelon form

Apply RJ ➔ ~ and ~ ➔ ~ then the above matrix reduces to,

1 0 3 0 2 I •

0 1 0 -3 3
0 0 1 -4/3 7/3
0 0 0 1 -2
 Linear Equations In Linear Algebra m l§w:r#a ll 1 ll
Apply R2 ➔ R2 + 3Rt and RJ ➔ o_ ~
...'3 + 4 Rt then the above matrix reduces to

-[~ ! ~ ~ 516]
0 0 0 1 -2
Apply R1 ➔ R1 -3RJ then the matrix reduces to,

0 0
-[~ ! ~ ~ 0 1
-5=::]
-2
is a reduced echelon form.

~ =i ~~ =~ -5]
4
8
Example 10: Given the mabix [ 6 ,1: ,discuss the forward phase and
. L
Pivot column
backward phase of the row reduction algorithm. [Model, CSIT, TU]
Solution:

o 3 -6 6 4
3 -7 8 -5 8 ,. .,
[ 3 -9 1~ -?.. 6 _,
)

; .j ... : • I
L Pivot column.. , I I j

Interchange rows 1 and 3. (we could have interchanged rows 1 and-2 instead.).
-~.
r~L9
i--:- Pivot .

-l!l
• I , 1·;.
J '

12 -9 6 I f .•I

~ 3 -7 8 _-5 . 8.
· 0 3 -6 6 4 -5 I • • ,,1

~
[3 9 12 £Pivot

O 2 · -4
-9
4
6
2 ~~]
O . 3 -6 6 4 -5
m
.-- Pivot
(g/¾<)_$c) l
"t '" ,:i

2 -6
4 -5
L New pivot column ..
6.: · 15]
2 -6
1 4
 MATREMATlcs-11
,..
-9 12 -9 62 -615]
-[! 2
0
-4
0
4
0
lL Pivot

12 -9
-•:-9]
-9 0

-n 2 -4
0 0
4
0
0
1

-n
-9 12 -9
I -2 2
0 0 0
0
0 -7
4
-9]
1

-[i
0 -6
I -2
9
2
0
0. -72]
-7
0 0 0 1 4

-24] '
-n 0 -2
I -2
0 0
3
2
0
0
0
I
-7
4
This is the reduced echelon form of the original ma~.
') i.
t
)•

The combination of steps 1-4 is called the forward phase of the row reduction algorithm. Step 5,
which produces the unique reduced echelon form, is called the backward phase.
Solution of Linear System. '1h u - r

The row reduction algorithm leads to the solution set of a linear system when the algorithm is
applied to the augmented matrix of the system.
Suppose, for example, that the augmented matrix of a liriear system has been changed .into the
equivalent reduced echelon form

[~ ~ -i
0 0 0
!]
0
There are three variables because the augmented matrix has four columns. Here, 1st and 2nd column
are pivot column. So the variables x1 and x2 in the matrix are basic variables and the 3rd column is not
a pivot column so XJ is the free variable.
The associated system of equations is
X1 -5X3 =1
X2 + X3 = 4
O=O
Hence general solution is
x1 =_1 + Sx3
X2 = 4- X3
{
X3 is free
 Unear Equationa in Linear Algebra W l§w:ra ij
Example 11: Find the general solutio .
reduced to n of _the linear system whose augmented matrix· has been

~
0
-2 3 0 -24]
ill -2 2 O -7
0 0 0 QI 4 _
Solution:
Here, has 6- column, so havin 5 . bl , · ·' '
d fr . g vana ~s. Among them, xi, x2'and x5 are basic variables and X3
~ t~ arerd eedvanthables because column 1st,· 2nd ~d 5th are pivot column and there has n o .
( t p1vo m 3 an 4 column. .
The associated system is, .- . ' '. 1 . ) ,

~1 ~ 2x3 + 3x,., = -24. J

X2 - 2x3 + 2x, = -7
Xs= 4.
-t:
From above system we have,

. .
X1 =

X2
-24 + 2x3 - 3x,.
= -7 + 2x3- 2x,.
,.

X3 = free.
X4 = free.
Xs = 4 (fixed).
This is required general solution. _

Existence and Uniqueness ·Questions

In our study of linear system, we observed that sometimes, th~ linear ~ys~em co~ists free
variabl~ (or free variables) and so~etimes there is n one -such free v.ariable exist. H, the system has
4 , ' • ! - • - -- r

free variable then the system will satisfied by many differ~nt ~oJutions, (as different vafue of the free
variable). This means the solution have infinitely :many sqlutio~. On1the other hand, if a system has ·
no one free variable then the solution will unique.
The following theorem leads the concept:

Example t2: Illustrate by an example that a system of linear equation ~has. either no -~ol~tion or
exactly one solution. ·
Solution:
Consider linear equations x + y = 5 and x + Y = 3.

Augmented matrix 1 [1 1 5] [1 2 57
1 3 ~ 0 0 2.J ·
I ,I '
I MATHEMATICS-D

which shows system is inconsistence (because last row is a form of [O O b], where b
solution.
Consider the linear equations
~ 0) so has l\o

x+y=3
x-y=l
Augmented matrix is

.[! ·:1 !] -[~ !2 !2]

which shows system is consistence (because last row is not a form of [O · 0 • b], where b * 0) and X1

and x2 are basic variable but there has no free variables so has exactly one solution.

Example 13: Solve the following system of linear equations, if consistent.

2x-3y+7z=5
3x + y - 3z = 13 _
2x + 19y-47z = 32
Solution:
j.'
The augmented matrix of given system is,
-3 7
1 -3 13]
rn 19 -47 32

Applying R2 ➔ R2 -! R1 and RJ ➔ RJ -R1 then .

JI
-3 7
'1
11/2 -27/2 1!12]
rn 22 -54 27
. .
Applying R2 ➔ 2R2 then
. .r • .
[2
Q
-3 7 ~

lf " -27 . ·11]

• ,.., • I

.)

~ 0 22 • -54 27
,r
Applying RJ ➔ RJ -2R2 then
...3 7

rn
11
0
-27
0 f] . . . ..
Applying R1 ➔ Ri/2 and R2 ➔ R2/lt the11
1 . -3/2 7/2

I
[
0
O
1
O
-27/11
0 YJ
,~ . • 3
Applymg R1 ➔ R1 + 2R2 then

n
1 75/11 0
0 1 -27/11
1

[
0 0 0
Here, the last row is in form of [O O O b ], b -:;.: ~ . ~ ' system is inconsistent.
 Linear Equations in Linear Algebra W l§ii!R 11
Example 14: Determine the value of h ~~,;,
and k so that system of linear equation
X1 + 3x2 = 2
and 3x1+~=k

has infinite many solution.

Solution:
The Augmented matrix of given system is,
1 3
[3 h

[~ h 9 ~ =
k 6] is echelon form.
For infinite solution in the above echelon form'must has at least one free variable, so
h-9=0 ⇒ h=9
and· k-6=0 ⇒ k=6

Example 15: If one row in an echelon form of an. augmented matrix is [O O O 5] then the
associated linear, system is inconsistent. · ,
Solution: ·

Given that an augmented matrix has a row of the form: [O O O 5] then,

,_ the associated equation
to the IpW is,
-
0.x1 + 0.x2 + 0.x3 = 5.
f l

⇒ 0+0+0=5.
⇒ 0=5.
This is not possible. This means, the system has no solution. So, the linear system is inconsistent.

1 HXERCISU 1. 1
1. Solve the system by using row operations.
(i) x1 + Sx2 = i (ii) 2x1 + 4x2= -4
I l;
-2x1 -7x2 = -5
2. . d thepom· t of m·tersection of. the straight lines x1 - 5x2 = 1 and 3x1 - 7x2 = s; ,
Fm 1
. h al s of h such that the matrix is the augmented matrix of a consistent linear
3. Deternune t e v ue .
system t l

(i) [; ! ~] (ii) [13 -2h -23] .(m). [13 h6 4]8 · . 1

.
so that the system has (a) no solution (b) a unique solution and
4. Find the value o f h and k .
(c) many solutions.for the system of equation. ' •1 .Jl.1 , '
.)
(1 hx _
X1 + 2-
. 2 (11.. ) Xt + 3X2 -- 2 (iii) 2xi - X2 = h
3x1 + hx2 = k
 I
MATHEMATICS-II ·
h e augmented matrices are:
. d the general solutions of the systems w os ]
5
· Fm
[o • s
1 3 4 71
(i) [ 3 9 7 6J
(ii)
1 -6
1 -2 7 -6
-
t.3

f !] o~:l I ntl
3 -4 2
(iii) rn ~ (iv) [ ~ ~ - =! J l.
not.
ofv

6. Determine if the system is consistent. Ve

(i) Xt + 3X3 = 2 (ii) X2 - 8X3 = 8

(iii) X1 - 2x2 + X3 =0
. 2x1 - 3x2 + 2x3 = 1 2.x1 - 8x3 =.B
X2-3X4 = 3
Sx1 - 8x2 + 7x3 = 1 -4x1 + Sx2 + 9x3 = 1
-2x2 + 3x3 + 2x4 = 1
3x1 + 7x. = -5
(iv) X2 - 4.x3 =8
2x1 - 3x2 + 2x3 = 1 -2.x1 + 4.x2 + 5X3 - 5X4 = 3
Sx1 - 8x2 + 7x3 = 1 3x1 - 6x2 -: 6x3 + 8x4 = 2

7. Test the consistency of the following system and solve if possible: 1 :

E
(i) 2x + 3y + 4z = 20 (ii) Xt - 3X3 = 8 (iii)" X + y + Z = 4 [2075]
3x + 4y +Si = 26
3x +- Sy + 6z = 31
,.. ' 1 2x1 + 2x2 + ~x3·= T ~
x2 + 5x3 = -2.
x+ 2y + 2z ~ 2
,,
2x + 2y + z = 5
.
8. Do three lines x1 - 4x2 = 1, 2x1 - x2 = -3 and - x1 - 3x2 = 4h~v~ a c~mmon point in intersecti,;m?
Explain.
- '1

r , 1 ' r,

(i) (-8, 3) (ii) (12, -7)

3. (i) h=7/2 (~) h*-6

- (iii) h *2
4. (i) (a) h = 2 and k ;t 8 (b) h ;t 2 ,, . (c) h=2andk=8
(ii) (a) h = 9, k ;t 6 (b) p~9 ii (c) h=9 .a n4k=6

f''
(iii) _(a) 3h + k ;t 0

=-5-3x, I- r•
(b) not possible,

=.t -t; Sx,

(c) 3h+k=0

r=.-17/2

t,
5. (i) X2 is free (ii) X2 = 5 + 6X3 (iii)_ X2 = 1/ 2
X3 = 3 , I '- X3 is'free 1
X3=Q
=4/3x,-2/3x,
_(iv) X2 is free ·'
X3 is free

' 6. •(i) Consistent , .r ' (ii) Consistent

C
J J
(iii) Consistent
(iv) Inconsistent (v) Consistent (vi) Inconsistent ' ,
7. (i) (x, y, z) =(1, 2,-3) (ii) (5, 3, :..1) (iii) (6, -5, 3) ' · · ·
8. One point in.common.
 Linear Bqaatlona In Uwr AJ&ebra W l§wja ij
J.3 VECl'OR EQIJADONS
.J ,•r ,.,. I

In the study of linear system, the position of notation plays a vital role. Sometimes the position of
notation may able to affect the solution. The vector is strict in the sense to determination of position
of variables and notation. This means the idea of vector is useful in the study of linear system.
Vectors in R 2

1Exampl~ 1: (i) ~ 1r [~] ~ii) . v = [ -;]

. ·,
(iii) w = [1 ~2].
H ~,:e (i) apd (ii)~ column vectqr~ and (iii) is a row vector.

Example 2: (i) U '= [:8 ,and y _=;[:~,. (ii), .. ~= [~ and v = [~] .

I• < ••
Here in (i), u = v if u1 = v1 and u2 = v2. ..I

But in (ii), u and v are not equal because 3 2 however 2 = ~-: *

Sum of Two Vectors
• • of two vec tors m
The addition · R2 means the sum of corresponding entries of the vectors.
For an instance if

[.j<t
.
,j~ 'Jj' j , l
-.· U, [Ul).ap!1
uJ V=[:~ , y I, , ,..

then sum of u and vis defined as, -[uil [vil = [ui + v1l.
u + v - uJ + vJ u2 + vJ
.J

f27· Th. ~n
f3J· and ,v = L2J u +v =[ 37 +[~ :[~: ;] =[~] ·
Example 3: (i) If u = L2 2.J

(ii) ff u =[ ~J =[~]v ~computeu +~v, u-~-

Solution: I •

Here,

('
 MATJIBIIATICS-0

Scalar Multiple of a Vector: . f by c is denoted by cu at\d . Note:-

il ultiplication o u , ., . ts augm
If u = [:Jin R2 and c be a scalar the~ the m , , . .r ~ . . . ~ i • • ·, , + ...+
• ,• 'di r1 •
.defined as ,. •_rl •'

cu=c[:~ =[:~.,y-~ · r ,:,; ·" ·• ; ·

· Example 4: If u {i] and c ~ 11 then cu= 11 [ 7] {:11. ., ,

Geo~etric descriptions of ·a 2 and R0 · : ~· · • d • rmirt d by

. . . h . t in the plane 1s ete e
Consider a rectangular coordinate system m a plane. Smee ea<; .pom
• 1 . [a] . 2 : 101 ; __, t in R2.' Since the vector
an ordered pairs of numbers. This means the vector b m .K 1s a P ~ ~ _ , ..... ·•

en~ fil'
representation is arbitrary in R2. This means the vector in R 2 is°a set ofall'point~ in'•tHe pl~ e.

the vector R' is the set of all points in space bec:use R

:i:COJ\~ •~ :•~· three
As similar, the vector Rn is -~ e set •of a\l ·points m: ~-din).~qsional space ,whei:e l!arull vector
represented as ··- · ·
.r\• ~; ' L;' :1L (,' ~ , • .. "/ • .. ::, .'r: "'1!•i·

u U1] .
= tii . ..,., . ·.·•· . . . .' f! 'J .. j ) f .!,; • )

' .
[...
Un :· .,-, ,,1 t

'10':l.
Parallelogram Rule for Addition
• l ,
ff u, V in R2 represents points in a plane then U + \ , represents the fourth vertex of the parallelogram
whose three vertex are u, v, 0. · • ,, •·.. •·. _. , ;:,,. · . · :, ,ru

Algebraic properties of Rn '

... I' r •

For all u, v and w in Rn and all scalars•c an4 d,t

~
• I
; .' . ;{. i . ' ~-
(i) u+v=v+u
t
(ii) c(u + v) = c\i.+ cv ·
(iii) (u + v) + w = u + (v + w) (iv) (c + d)u =cu+ du
(v) u+O=O+u=u ·(vi) ·' c(du)° :i:: d (cu) = (cd)u
(vii) u + (-u) = u - u = 0 (viii) 1.u =u.l =u.
 .Linear Equatlona In Linear AJp1n W ICaAP§ 11
Note: A vector y e Rn is linear com.bina .
. tion of vectors .
augmen ted matrix [u1 u 2 • • • U1, u2, U3, • • • Un m Rn, if linear system with
- . .
+ ... + XnUn - y ts conststence.
Un ·Y] represents ...- . .
cunststence system i.e. equation x1u 1 + x2u 2

by . + CnUn.
@Mi A vector b is in span {
[u'i u 2 •... Un b] rep ::.11 u2, : · ., Un} if linear sys~em with augmented matrix
ittW resent e conststenc~ system. ,
(i) span {v1, v2, ..., Vn} contain each v· i = 1 2 , bee
v· = Ov1 + Ov + + ., , , · • ., n ause
I 2 ••• 1Vi+ 0v
(ii) spanQv{V1, V2, . .., Vn} c~nta~-~ron~ector also, because
0 = 1 + Ov2 + ... + Ovn

:or
Example 5: Let u = [ -1 l ] andv -[2
- 1], show that [h] . .the Span {u, v) for .all hand k.
k 1s
1

Solution: - I ,-

Given,

The augmented matrix form of u, v and b is,

m lu v b] ! ['2 2 kh]
= ~1- ·1 ·

~[~ ~ 2k:h] [° .. applying R2➔ 2R2 + R1]

l . R2➔ Ri/2
~ h/2 ] ..
~[~ 2k+h/4
[ · app ymg R2➔ R2/ 4]

0
1
'[f
~ 0
(h- 2k)/4]
[": applying R1➔R1-R2]
(h + 2k)/4
This shows that the solution of the augmented matrix exists for all value of hand k. So, bis

- ·
Span {u, v}.
' I I~ - . '

If the augmented matrix form of linear combination is inconsistent then the vector b does not
span the given vectors. .. 1

f ~ •, • 1. '
Geometric Description of Span, {v} and Span {u, v} Iin IR3
Let v be a non-zero vector in R 3 then Span {v} is the set of all vectors in the form

CtV
... (i)
I '. I(. • I

with c1 is a scalar.
 MATHEMATICS-D
· dim · n v So it represents a line through
Oearly -the form (i) is linear and has one ens10 . . . , . R 3 t .......... g o . . Ori~
· . f Span {vJ 1s a line m con a~......... ngm.
Therefore, the geometrical representation ° ,
• R3 then Span {u v} be a set of all vectors in the fotn\
As similar, if u, v are non-zero vectors m '
... (ii)

with c1, c2 are scalars.

Oearly, the form (ii) is two dimensional. Therefore, the equation (ii) represents a plane
containing the origin.

. _ . . · [1] [-!;S]
Example 6: Let u • -; , v • [-3]
and b '!' -~ , Then,is b in the plane of span lu, v)?

Solution:
Given,

u=[Hv=[-J;] b=[H
I
~ d
''-· I! .
Then, the augmented matrix form of u, v and b is,
5
[u V b]=[~ -13
-3 I] 1,,
• I • J
• •
I..,
l

5 -3] ,. ' '

- R2➔ R2+2R1

-rn -13
-18
io [". applyin~ &➔-&-3R1 ]

·[1o
~ 0
5
-3
0
1)
-2
(' .· apply~g & ➔ & -6R2]

This shows that [u v b]- is inconsistent. This me~ b does not li~s on Span \u, v\. So, bis n
in Span {u, vJ. That is·b is not in the plarle of Span {u, v}.
1 -2 4

U]
' I .• I

[!].
l • ' l • • . • ".' I • .

Example 7: Let a, =[_~] a2 = and b = For what value of h is b in the plane spanned by-

and a2?
Solution:
We have if x1a1 + x2a2 = b has solution, then b is in the plane spanned by a1 and a2•
• I •

Let bis in the plane spanned by a1 and a2. ~' the equation

Ax=b

has solutic~I).for A = fa1. ,.a2]

 Linear Bquatlom In Uneu Alpbm m larAma 11
Here the augmented matrix of Ax. bis - -

[-2! :;7 1]h

Apply R2 ➔ R2 -4R1 and I¼ ➔ I¼ + 2R1 then ,, I I •

2
[~
0
~
3
-is]
h+8
Again apply R2 ➔ R2/S and I¼ ➔ !¼/3. Then

[g -112 _; ]
(h+8)/3
Again apply I¼ ➔ I¼ -R2 then

1 -2
0 1 '43 ]
[
0 0 (h+17)/3
Clearly, the .matrix gives solution o~ if

h+17
3 -0.

⇒ h=-17.
• I

Thus, for h = -17, the vector b is in ~e plane spanned by a1 and a2•

' f

1.4 THE MATRIX EQUATION Ax= B

ot

1 -The product Ax is possible only if the total number entries of A and x ar~ same.
Example 8: For ~1, v2, v3 are in R3, write the ~inear combination 3v1 - _Svi + 7v3 as matrix form Ax.

Solution:

Here,

I ('
 MATHEIIATICS-D
~ f lin uation as the form Ax = b. ,
Example 9: Rewrite the following system o ear eq
Xt + 2x2 - X3 = 4

-Sx2 + 3x3 = 1.
Solution:
Given system of linear equations be
Xt + 2x2 - X3 =4
-Sx2 + 3x3 = 1.
This can be written as,

Sc

⇒ Ax=b.

where, A=[~ -5
2
-JJ,
x = [;~ and b = [f].
The concept of the example leads by the following theorem.
'naeorem 3: If A is.,an m }( n m~ ·
Ax= b has the sam ·

which in ·
1Il;8l:rixis

Importance of the Theorem:

,
The theorem pointed that a problem can be represented in three ways
(i) as a matrix equation, • A \, .
(ii) as a vector equation and
. (iii) as a system of linear equation. This means we can switch from one formulation of a
problem to another whenever it is convenient.

Existence of Solutions

The equation Ax = b has a solution if and only if b is a linear combination of the columns of A.

Note that if the augmented matrix in echelon form have a row as the form [O O O b} (with b * O) then' the
system is inconsistent being the last most column is an pivot column.

- T h e theorem (d) means only for coefficient matrix A but not for the augment matrix of Ax• b.
 Linear Bqaatlom In LIDeU' t:11 lira Ill fca½# ii
. [1.:1 3 0
,
Example 10: Let A = ~
-1
-:
-1
: ..
(i) Does the equation Ax = i; has a solution for each be R'.
'lf
.'
(ii) Does column of A span R•.

(ill) Can each b e R • is linear combination of column of A.

Solution:

All above equations answer will give by knowing A has pivot position in every row or not?
Here, l_
I

A=
[L'3
-1
0
-1
-4
0
-1
2 -8 i] "l ,..,_

r
2 0 3 !
-1 Ir

By elementary row operation, we reduced it into echelon form

m
3 0
J
43 .:- . . ., - . ~

{!
a c _ -, ~ -!
-1 0 . -. ' -· .< ~ • ~ r- ;- , .
I - (. ,,
0 0 [fil . ,. .
I I
.-1.1,.
• •
•
-•
f" .I~
•
-:-1 I
.. r'
0 0

So having pivot in 1st, 2nd and 4th row, but not tri 3rd row. Thus -
. I I..; . '-'. -== I 1, - • • I - •

(i) ~ = b 1!as no solution for ,each b e R 4· 1... • t . J

(ii) Column of A doesn't sp~ ~ 4 • •

(iii) Each be R4 can't be the linear combination of col~ of A.

Solution:

Here,

[Applying R1 H R3]

This means {v1, v2, v3} Spanned R3.

' '
 MATBBMATICS-11

Example 1~ la A•[-: ! !l v1 • [:]. v, +:] and c • 5. Show_~ t

• I
Exl
(i) A(v1 + v2) = Av1 + Av2.
Sob
(ii) A(cv1) • c(~v1).
Solution:
Here,

and,

Also,

Then,

Now, (i)

(ii)

[-23J =r-115J
.'.i. ' (It:
c(Av1) = 5 42 210 . · ·' ·
-61 -305
and

=rg~:g;~!l
l~ + 1s-32~J
=[;;~J
-305
= c(Av1).
 Unear Equatlona In Uneu Algebra m ICRAfflR i)
Example 13: Solve the mahix eq

Solution:
. [ 1
uation Ax = b where A = -3
O
~5 !J,b=[~]
3 -1 ·
Here,

[1
-3 . 12
1
[A: b] =
0 5
2
3 _t]
2 1

-rn 7
5
2
5
3
1
I] [applying R2➔R2 + 3R1]

-rn 7
0
5
-4
_t] [applying R3➔7RJ - SR2]
2 1

-rn 7
0
5
1 !] ..L, ·./~
[applying R3➔R3/(-4)]

2
0 -3]
-rn 7
0
0
r '3
-14 [applying R2➔R2 - SR3, R1➔R1 - ~]

o ,_
3] . "... •_; t;;il ' - i/.
, II •:, .
2 I

-rn 1
0
0 -2
1 3
[applying R2➔R2 /7]
" l f lj r

r [1 0-~ 1].
- 0 1 -0. --2 . i!
,· ,,
-
[applying R1➔R1 -2R2]1
I I

. 0 ·O 1 3 I •

This shows -the solution of the matrix equation is (1, -2, 3) .

'I - .
...i._BXEBCISE 1.2 I - I
.
1. Compute (u + v) and (3u - 2v) when

(i) u•G] andv=[tJ ) ., • l

2. Determine if b is a linear combination of a1, a2 and c13. 1:-1

(i) a, m[¾], a2 s(!Jaa~

[:J ~·=[il '(;): a,=[-}] ,a,=Ul aa~.[il ~=l~l
, I - I

3. Determine if b is a linear combination of the vectors formed the columns of the matrix A.,~ .

~
, (ii) A= 0
[1 -2 -6] =[11]
(i) A= [
-2
1 -23 57 , b -59 .
 MATHEMATICS-ll

4. Find the value of h so thai vector b is in Span {u, v} wheie •ti =[ !] ' v= [!lb [tl · =

5. For what value of h with y be in Span(v1,

. V2J
} h
v 3 w ere Vt -:-
- [1]2 ' ,[5]
-1 V2 = -4 ,
-7
V3 = [-3]1
0 at\d
The
solu

y=[tl 7
(2076]

6. u,tA=[!-3
3
2
-2
!]
-7
and b = [. ~] . Is. the equation.Ax = b consistent for aµ possible bi, b2, hJ?
~ . . . .
7. Write the following system as in matrix equation N
(i) 3x1 + X2 - 5X3 = 9 (ii) 8X1 - X2 =4 0:
'· ,
X2 + 4X) = 0 Sx1 + 4x2 = 1 E
X1-3X2 = 2

4l r2J
r
• : .. ,.. (.I r .. '

~
2
8. Solve the matrix equation Ax = b where A = [ 1 ~3 , b =. ~- . .
-2 -4
; . 8• , .
9. Is u in R3 spanned by the columns of A? j

,~I, ! (i)
u=[:JA= H-5] ~ :.i_ . •,r ' t (ii) [2r ~5 ··7]
2 ~ t
,:
u=-3 .A= JO 1
:
:1 3
,8

i · I
,· · !."' :ur, .t j-. _-, .. l , } ;',.,_"}' ,(Jr: ; .. -.,., I

1. (i) [!],[~]
2. (i) b is a linear combination of a1, a2 and ~- ( J j •• .. ! I •
;r .
(ii) b is a linear combination of a1, a2 and ~- .- , :
3. (i) bis not a linear combination bf columns of A.
(ii) bis a linear combination of columns of A: ...
4. bis Span{u, v} when h = 3.9. " . • I ,r
I I j

5. h=S (
I , r 6. lnconsisten~for every b·
I r
I

7. (i) ri: -Sj[~-[9J0

1
1 4
J
X2 =

;
!

I
(ii)
[~ 1Jt.:~ ~
rnI

8.
[1] I f"
9. (i) Yes
(ii) Not.
''
 Linear Equatlona In Uneu r----
( .S SOLIJflON SETS OF LINE _,_ ID IC,,AmR tJ
AR SYSTEMS

11te solutions of linear systems are . •

solutions which we discussed below IDlportant objects in linear algebra • A systern has clifferent types of

.
Note that,
one free the equation
variable. · . . salution if and only if ~ equation has. at,, least
· will have such a non-trivial

Example 1: Determine if the system has a trivial soluti~n

Xt - 3X2 + 7X3 = 0

-2x1 + X2 - 4.x3 = 0
Xt + 2x2 + 9X3 = 0.

Solution: Given system of linear equatio~ be

X1 - 3x2 + 7X3 = 0

-2x1 + X2 - 4,x3 = ()
Xt + 2x2 + 9x3 ·= 0.

The augmented matrix of the linear syste~ is

I 1' . - '
·[!2 -; : ~]
°'., 1
~
2 ' 9 • 0 ' •' •
[t
0
=!.5 {o2
• f -3 7· 1.
f I

el 10
0 rg]
The pivot are in 1st, 2nd and 3rd column. So, x1, x, and x, are basic variables and having no free
variables. Thus the system has trivial solution. · ·• · · ·• 1
.:. ' 1

~ple'2: l)etenlUlle if the system has a non-trivial solution.

.. ' 2x - Sx1 + 8.i:, = 0 , .
1
-2x1 - 7x1 + X3 = O • 1 ' :J
4x1 +2x1+7x3=0.
J • • •

Solution: The augmented[m;irix_;t ~v~Jsystem of linear equation 1s,

.. . r
J ••

-2 -7 1 0
4 2 7 0
2 -5 8 -'.
,.., 0 -12 9
[ O 12 -9
 IIATBBMATlCB-11

2 --512 80]
~
[0 0 90 00 (Apply&. ➔ &, + R,
o 2]

,.
! ~]
I. '
I Exa

-rn : [Apply R, ➔ R,/3]

-rn
~
r 0
8
-3/4

rw ~ :;~: ~]
~] [Apply R,

[Apply Rt
➔ R,/-4]
➔ Rt + SR2]
Sc

L~ o o o .
Here, pivot column are 1st and 2nd, so basic variable are Xt and x2 and X3 is a free variable. So,
the given system Ax = 0 has non-trivial solution._
The last matrix gives,
I '
17 17
2xt + 4

X2
=0

3
⇒ X3
<i X3 =0 ⇒
Xt=-gX3.
X2
3
'

= 4 X3 .
~· t'c •

0 =0 ⇒ X3 = free. f =

Thus, the system has the general solution of Ax = 0 in the form

x=[:;J=[-!::i~ ]·=x3[-!;~ ]=x,v: :: "•·

8 8
. . , ,. ;
~ ~ 1 I\

Hence, the given syste~ has non-trivial solution x = v. X3

~ .:t., i[':dL.
Note: In this case, the solution is set is Sp~ {v}: -:-
• • • I
! • l !" • l
ff the equation ·has only one fr~ vatjable, then the solution set is a line through origin. If the
equation has two free variables then the solution 4; a .pl~e containing origin. Likewise the solution
has different geometry p5 the number, of free yariab~ involve in the equation. Also, if the equation
Ax = ohas only Oas its solution then th~ s6lu~~n seti s Span {O} and it is a point origin.
H1.
Parametric yector Equation: •. I
• I

In above example, we observe the solution pf the linear

. system is in x = x3v (for ,x3 is free)
. then
the form x = tv (for te R) is called paramebic vecto~ equation and vector notation of this
solution is called parametric vector form.
H the linear systems has non-trivial solution as x = X2u + X3V (when two variables x2 and x3 are free)
then the parametric equation is x = su + tv (for s, te R).

Sol,u tions of Non.:..homogeneous Systems

A liner systems of the form Ax = b is a non-homogeneous lineM systems. The solution of such
non-homogeneous _systems can be written in P.a rame~c vector form as a single vector plus a
vector form that satisfy the corresponding homogeneous system.
 . . Uneu Bquatlona In Uneu r = ~ ~...
Example 3: Determine the soluti . M&i6ii m ICaAPna ii
on lll paramebic vector fo f th
x + 12 ~ rm O e following linear systems.
I ~2 +x3=t

-4x1 -9x2 + 2,x3 = -1

•
-3x2 - 6%3 = -3.
Solution:
The augmented matrix form of th . ,. i , , ., • I,

[!0 -}9 ~
.•,,.
-~] e gtve[: lin';" •~lis], '
3
[~pplyR2 ➔ R2+4R1]
{ I
-3 -6 -3 O 6 3·
0 -3 -6 -3

[~-!0 ! llqJ
0 O
[Apply RJ ➔ RJ + R'2] ·
. (.

[o( ( i ' ~]
0 ·0 0
So system is a consisten~
·
Here, 1st and 2nd column have . . rrr
a pivot so X1 and x2 are basic and X3 is a free -variable. .,

- 1. [
0
1 0 -5
1.
-2] 2 1
' i ~ Q ()l .. Q I# Q' .(" ,f T'
This gives, Sx3 =:.2 · ⇒ xi = -2 + 5~3
X1 - l - . -
X2 + 2x3 ::·1 . ;=>:.f2 = 1 -2x3 ')
r . . •
X3 is a free
Therefore, the gener.@.! solutiqn qf the.system has the form

X = [=:( [-;_+~ ~ ]; [j2] . 3

] [~
2 Xa
· X~J X3 • Q t · 1 ::
r.
⇒ x=p+X3V
ForteR, X = p +' tv• . · .
is solution of given system on parametric vector form.
. '- I
1
Note that, the vector x = p is the particuh\r solution of ilie. linear system and x = tv be1 solution of
homogeneous part of given linear system. ! · ·
In above example, we obse~e that the system ~ = 0 has solution ,x = tv and the system
Ax = b has solution x = p + tv. This means if the 1 equation has effect b then the solution x = tv
translated by p to x = p + tv. · r ~ ·

Thus, the relation between the solution of Ax = b and Ax = 0 is the translation. ·

1
The following theorem gives the precise statemen~.

. ::iMlm~-) ~. onsistent ·'°' .

ofAx
fflfi1• the equation Ax= b has no solution then the solution set is empty. This means there will not
be any translated value.
 r
lt_l~ERCISE 1.3
· d ·te the solution in parametric fonn .
-•••
We
sol,
1. Solve the following system of linear equations an wn , if
( I• I .,

possible. , •,

(ii) +y +Z = 6 (iii) X + }' - Z =: 9

(i) 6x +4y = 2 X

3x-5y =-34 .
x-y+z=5 .
3x + y + z = 8
By+ ~z = -6
-2x + 4y -' 6z = 40

(iv) 4y +. 3z = 8 (v) 7-x - 4y - 2z = -6 (vi) 2x1 + Sx2 + 6X3 = 13

I I
2x-z=2 16x + 2y + z = 3. 3X1 + X2 - 4X3 = 0

3x+2y=5 X1 -3x2-8X3 = -10.

(viii) 3x + Sy - 4z = 7
-3x-2y+4z =-1 • I J
, ..
6x+y-8z =-4 I - I I
' (" r !I
2. Determine, if the following homogeneous system has a non-,,trivial solution.
(i) 3x1 + Sx2 - 4X3 = 0 . (ii), _ ~1 + 3x2 ~ 5~ =_O :. r1
- 3Xt -2x2 + 4X3 = 0 Xt +.4,s2,- 8X3 = 0 ~ • .. J • •

1. i. X =-3, y =5

iv. Inconsistence, no solution.

J ' ., ; J ~ I t

•l
vii"
H

2.
l '

. '
' .
 t1neu Bqaatiom III t1neu Algebra wlea= 11

J
if
J.6 APPU~ATIONS OF 1..INEAR hnEM

We expect a real life problem in.volvin lin

solution. But sometimes the problem m! h ~ :U~~a would have only one solution or perhaps no
Y a~e infinite solution.
Example 1: Suppose an economy cons· ts Of th
output of each sector•is dis;b , ~ ,Coal, ~ledric (power), and Steel sectors, and the
the entries in a colun\n re uted among the various sectors as shown in table, where
pr~8 ent the fractional parts of a sector's total output.
The second column of T bl .
1
, : - 10• •'1sector is divided as foll~~~ for , ~tanc-=, says that the total output of the Electric
Electric. Since all ou ut · ~/o to Coal, ·500/o fto ~el, and tlfe remaining 100/o to
column must sum to~ must be taken into account, the decimal•fractions ·in reach
Denote the prices f d ti ' :-, 1 • . ra . ~ •,:, .r• r • , ~ ,, ,r
and Steel secto bi.e., 0 ar,values) ~f the ~total annual outputs of the ~oat, Electric,
that make each rs t! re:
PEt and P5, respectively. If rp o•ssible/ find equilibrium"prices
sec r s mcome ma~ch its expenditures.
' , I J . ·,
A sim le econom - · =\
,Distribution of Outpttt from
:. · '·CoaP .:·_ , .Electric "-, . . · ·
ft• JO, 1' ,.· . ,t' o~o -'-. 0.4 : Coal '.,
.., b :1
I
· · o.6 · .. 0.1 .. rt• •
0.2
I ,. . ~ •
.
i ,I
ElF.tric .
, 1 rj

t I

0.4 · 0.5 . 0.2 Steel

Solution:
A sector ldoks'down a columh''t:o 'see ·wher~ its oiitput 'goes,'and it looks a ~ -a row'Eo se&wruit it
neec:fs' as inpu'ts'. 'For instance, th~·fitst r6w bf, Taole says1that Coal 'receives·(aha pays' for) 40% of tne
Electric output and 6()%: of-the Steel oiltput.' Siri~e ~e respective'values of ilfE ·total ofrtp~1s arel)E.ahd
p.;, Coal must spend 0.4JlE· :dollars for its share of Elec~ic•s: output and 0.6ps for its share of Steel's
. ou,g,?,t. ~ 'm C~at'f t~tal e?'p~nses .~ e 0._4J?E ,+ O.?ps. To malfe ~~al•.~inc~~e:.Pe, Ejl~al to i~ px~nse~,
we. wantI If ',,
f J f , r

~~y, th~ third row of ~~- exchange ~ble leads t~ 1th~ ,fimµ r~uireme1?-t: )~, _ , ,. ! .
, c:,. ps ;:,0.4pc + O.SpE + 0.2ps ... (2) " )~ , , ,
To solve· tl\e system ·of equations (1), (2), and (3), .m~v~.all the -~ o ~ ~~' ~e•le~ ·s~~~ of th~
~uations and combine like terms. ·
pc - 0,4PE - 0.6ps r= O
. : . , - 0.6pc +o:9pE 7 0.2ps ~'o
. ' - . .- 0.4pc .: 0.5PE + 0.8ps = 0 .
. . . h d . : .....:-... 1., are rounded to 'two 'places.' ' · · r
Row reduction is next. For sunplic1ty ere, ecuuaw
1 O4 -0 6 OJ . , [1 -0.4 -0.6 OJ
-0:2 0 . ~ -0.9 0 0.66 -0.56, 0
[ -0.5 0.8 o~:! 0 -0.66 0.56 0
 IIA;1'JIBIIATICS-D

1 -0.4
0.66
-0.6
-0.56
~OJ
[~ 0 0
1 -0.4 -0.6 ~]
0 1 -0.85
[0 . 0 o , b ..,
'., j
1 , . O -0.94 OJ .
O 1 -0.85 0 .
[
• I Q O O Q .
• •
1
• • • • • • • • • • ce vector for the
The general iSOlution is Pc = 0.94ps, PE = 0.85ps, and ps js free. The equilibnUDl pn .
economy has the form · , , ·

The general solution is pc = 0.94ps, PE =0.85ps, and ps in free.

'"Jl - • f • 4 •I r ~ 1 I- , r 11 • J., , •J. • . •
, The equilibrium pti~.VectoJlfo:t: the economy has th~ form, .,, . 1 ,,• --.

o94 · ] [n>9tj:;J - 1

[ ] [o:ss: = ps o:ss . ,.
ri • :.~, • • · · ·
p= : =
p,;, ,~ ps I 1 , •o

- - - - # ~ ' • ' ' \. -- - •

J Any (nonn;iative) ~oice fo; ps~~~ in a choice oh?qtillibri~-prices. Eoninstance, if we take ps to

' be-100 (or $100 milli9n), then~-=
94 and PE= 85. ~e incomes and expe~~tures-of sector will each
be equal if the output of Coal is oriced at $94 million, that of Electric at $85 million,
1J
and that of Steel
I I f•
at$100million , - ' , · · · ·
f ' •J. • ' '.'1 •\ .fl I
Balancing ChemicafEquatfoiis -- · - - -· - - - --·- · · - · - · ·
·no ,
µtemi~ eguaµons ,4.esc;ri~ tli~ qµantitj.~ -~f substances consumed . ~d · p~uce~ , ~ chemiQll
~ o ~., fpr ~ e , ~ ~en prop~e ~ bums, the propape (yffa) combin~ with oxygen (Oi)J o
form ~~n 1d,io~ <fet(c;:Ch) ~d-:wa~r (H29), accordipg to·~ _e quation of1the form
;, 'o ,J • . • • .: , . ,.(xt)~ +(1'Cl)Oi:➔, («3)C~*, (x4)H~ I : ••• (4) . :
To 'milafice' this eq~ation, i ch~t'must firid ~ho~ numbers xi, ......,·X4 that the total numbers such
of carbon (q, hydrogen (H), and oxygen (0) atoms on the left match the corresponding numbers of
atoms on the right (because atoms are nei,t her destroyed nor created in the reaction).
A systematic method for balahcihg chemical,eq'.ub;tions is to set a vector equation that describes up
the numbers of atoms of each type present in a reaction.- ~e equation (4) involves three types of
atoms (carbon, hydrogen, and ·P,~gen), .construct~ yector 11\ R3 for each re reactant and. pi:od .
(4) that lli;ts the numbers of "atoms per molecule" as follows: · ·•. ~ m

c~=[~], Ch=[~]. ~Hl ~m-~~~·

To balance equation (4), the ccxjfici~ ~1,. ....1 X4 m~tJsatisfy
 Linear Bqaattona
To solve, move all the terms to the 1eft(chan. In Llwr AJpl,ra g;i ICaAmat 11

mx,rnr:rj:in[;]e=thu[~~ and
--

x, + fourth :vectors);

Row reduction of the augmented .

matrix for this
1 i
-l
.
-1
• eq~ation leads to the general solution
~J
. Xt =-
4 X4t x2 =4"'1
S X3 =4"'1
3 with x.dree . · . . . L· ·~·,.. ir.

Since the coefficients in a chenuc. al equatio ·

the x2 = 5, and X3 = 3. The balanced equation IS
. n must be integers, take X4 "" 4, m· w hich case x1 = 1,

Qis + ~ ➔ 3G:~ +.4H:i() .

The equation would also be bal ed . . : .- . ! I • • , ,.. r

purpo , h owever, chemists prefer

ancto uif' forbalexamP1e, -each coefficient were doubled. For most
seswhole numbers.
passible se a anced ·equation whose coefficients are the smallest

Example 2: The Network in figure shows the traffic: fl (. .

!1ay stree~r,i n ~pwntowq Baltimore durinow m ~ehicles per hour) over several one-
•s ' . • 8"?"°1 flow pattern for the netw:,,,k. g ' lypl<al early afternoon. the ~
5 to .r, . ' .( 100

vill
:eel
Calvert St. South 's..
t
N
LombardSt. B C
300
X4

.t2 . .r,
cal Pratt St. A D
300 -,..-----.;..;.;,;.-F--....--~--___.600
.r, ..
to I

500

Figure: Baltimore streets.

Solution:
Write equations that describe the flow, and then find the general solution of the 5Ystem. Label the
street intersedions (junctions) andthe. unknown flows in the branches, as shown in figure. Ai each
owou.
intersection, set the flow in equal to the fl t Flow out
1ntersectio~ Flow in
300 +. 500 = Xt + X2 .
A
x2+"4 = 300+X3
B
' I~:= X4 + X5
C 100+400
Xt +Xs
= 600
D
Also, the total flow into the network (500 + 300 + 100 + ~) eq~ total flow. out of the netw'?'k ~
(30o + x, + 600), which simplifies to"'= 400. ComJ>ine_this equation with a re;mangement of the first
four equations to obtain the following systeDI of equations: .· . -
 MATBBIIATICS-11 ( . . ,,,
+ X2 = 800 ' ..
Xt

x2 - X3 + X4 = 300

x.+Xs=500
Xt + Xs = 600
(ii'
X3 = 400

Row reduction of the associated augmented matrix leads to

Xt + Xs = 60()
,I • I • .,
X2-Xs = 200
X3 = 400
x. + Xs = 500
The general flow pattern for the network is described by

::::~:
{ X3 = 400
x.=500-Xs
Xs = free :.. ; ' . .
A negative .flowjin a network branch corresponds to ,flow_~ the dif~tJ.~h OPP,~~ite t~ '~ t shown_ on
~e model. Since the streets in this problem are one:-way, none of the variables here can·be negative.
This fact leads to ·certain limitations on the possible vaiues of the variables. For instance, Xs ~ 500
because x. cannot be negative. ·
j
I.7 LINEAR INDEPENDEN(;E

An equation of the form Ax = 0 may; have different · type of solutions. The vector form of
Ax = 0 gives either trivial or non-trivial solution for the linear system Ax = 0. If the solution x = tv is
trivial then the set of vectors of V is linear independent and V is dependent if the solution X = tv is
non-trivial.

Geomebical Nature of Linear Dependent and Independent ,

·Consider vectors

(i) . \;l ~ [;] 1 V2 = [;] .

. [6]2 = [3]1 =
Here, v2 = 2 2
V1.
This . .
means the given vectors are linearly dependent.
,
 Linear Equattona In Linear Algebra Ill fciwija, ii
th
Geometrically, e vectors Vt and V2 are the position vectors of two points A and B whose
3
coordinates are ( , l) and (6, 2) respectively.· Oearly the vectors (i.e. straight lines) have same
initial Point 0(0, 0) and moves along with (i.e. collinear). ,

(ii) Vt = [~], V2 = [~].

Here, v, = [~ = 2 [n "'2v,. This means the given vectors~ ~arty ~~dent.

Geometrically, the vectors Vt and V2 are th~ position v~tors of two points A and B whose
coordinates are (3, 2) and (6, 2) respectively. Oearly the vectors (i.e. straight lines) have same
initial point 0(0, 0) and moves toward the point with differenfpath (i.e. not collinear).

(3, 2)

non
ti.Ve.
: 500

-
l of
:vis
Vis

Example 3: wtv, =[~], sm, m

3
v, v, =
. . · dependent or independent.
(a) Determine the set fvi, V7., v3I 18 linear1y . th
t f" d the relation between em.
(b) If the vectors are dependen ' in

Solution: · t equation of given vectors is,

(a) The augmented matrix form of the vec or
12 5 4 . 2, 0] 1

[3 6 1
0 0
0

4 2 R2➔ R2-2R1
-3 · -3 [Applying&➔ & - 3R1
-rn -6 -6
I IIATHBIIATICS-D

-rn
4
-3
0
2
-3
0 ~] [Applying R3 ➔ R3 - 2R
2

4 2

-rn 1
0
1
0 ~]
-[~ ~ ~2 ~] [Applying Rt ➔ R - 4R2]
0 0 0 0 1
, . en set of vectors are linearly
Here, X3 is free variable, so having non-trivial solution. Hence, giv ·
dependent.
(b) For relation between them

X1 - 2x3 = O· ⇒ Xt = 2x3

0 =0 ⇒ X3 =-free.-
·--
This shows that the given vectors are ~early dependent being X3 is free.

As a particular, for X3 = i, then we get X2 =1 and Xt = 2.

Then, the relation between vi, v2, V3 is

Solution:
-· ·The augment matrix form 'of the vector equation of {v1, vu v3} is,

7 9 7 9

[~ 2
-6
4
-8 i] -rn 2
0
4
4 i] [Apply~ ➔ ~+ 3R2}

7 9

-rn 1
0
2
4 i]. I • • ~➔ ~/4
[Apply.R2 ➔_ R2/2

7 9

-H 1 0
0 1

0
ff [Apply R2 ➔ R~ 2Ra]

9 0 OJ
;-
-rn 1 0
. '
0 1 0
•
[Apply~R1 ➔ R1 - ~]
 Linear Bquattom In Linear Alpbm W ICaAma 11
0 0

-rn 1
0
0
1 i] [Apply R1 ➔ R1 - 9RJ]

~]
0
L
--rn 1
0
0
1
[Apply R1 ➔ Ri/5]

Here has no free variables. This implies X1 = 0, X2 = 0, x3 = 0. This means the system has only the
trivial solution and the set.of vectors are linearly independent.
I,lnearly Indepe~dence of llatrlz Column
Let A = [ a1 a2 · · · an] be a matrix. Then the equation Ax = 0 can be written as

1 + X2 a2 + ... + Xn an = 0.
X1 a1
' I I

From this; we observe that each column of A corresponds to a non-trivial solution of Ax = 0

.
whenever the columns defermine linear dependence relation to the column.
.., , ' ,
• The ~i~ of a matrix· A ~ linearly inde~d~t if and ~nly if the equatio? Ax = 0 has only
trivial solution. That is, the columns of a matrix A are linearly independent if and only if the
echelon form of A has no free variables._. ,

Example 5: .D etermine if the col~ ~i"th~ ~ trix A=

9
r~
Ls~is . o 1

!1] ~e linearly independent.

. .
r• : , } r 1! 1 \_
1
-:;: j I l ,I Ji' •

Solution:
- I

The a~gmen~ matrix of Ax = 0 be, ,

f .. a .
.. & ,, ") ' -

[! ! -} ~] ''. .'. ·' ·.' r• r l .a,

lr(
•' •• :-a•
·: ·-.,.

2 -1
1 4 ~] [Interchanging 1st and 2nd row and en "apply Ri-'+ R,- SR,]
-2 ,5
l I t, I ·;

-[T ~ ,!1 ~] [~pply~➔~+~) . ~ -·:·:· : ,, .. _.

: . . free variable. So there exist trivial solu~ on. This

Here, x1, x2ffld X3 are basic vanable and has no . ,
means the columns of A are linearly independ_ent. .

Characterization of Linearly Dependent Seta . { l~ lin

· early

if - 0 This means, the set Vt 15

(a) A set of a vector {v1} is linearly dependent Vt - •
independent if and only if Vt ~ 0.
 MATJIEMATJCS-D if t least one of the vect ors is
. a lll.ulti
. . ly dependent ~ Pl,
(b) A set of two vectors (vi, v2} IS linear if and only if neither of the vectors is a ll\'U} . Ill
the othe; ' And, the set is lihearly independent tip~

of the other. . linearly depend ent if and o nly if at l

£,twO or more vectors ,1s ~
(c) A set of vectc?rs {vi, ..., Vn} ° • • f the others. But note that, not all Vect0
one of the vector of the set, is a linear combmatton
. • 0 f other vectors. ·" · .-
° .,, ·
ts
necessarily should have linearly combmatio~ - ! ' •

Proof: Let A= (vi, ... 'v.J with Rm. Then A is ·m xn matrix and its corresponding equation is
Vj E

A; = 0. Let n>m. Then there must be more -yariabJes µtan equatj.ons. Hence, on reduction of
echelon augmented matrix intq ech_elon f9rm. , 'Drl5 means ther~ m~st be a free _variable (at
least one free variables), so, Ax = Ohas a non-trivial solution. Therefore, the columns of A are
1,, ,
linearly dependent. t l . ..J • • r, J t '"). J

Ji Proof:.. Le~S = /~1, ..., Vn} in_

.
~m c~ntaiµs tn~·~ero v~tor~ SQ let v 1 = 0.
' . - '...-- \ ... •O

Assume that Xi= 0 for all i(~ 2, ... .:~>

nwith x '# othen
1

X1V1 + O.v2 + ...... + 0.vn = 0

This means the set is has non-trival solution, so the set ~£;ecto~ sis~lin
;. ear
. 1Y ~-
· aependent. < ._ •

Example 6: Determine if the given sets are linearly dependent or nol ... '· I

' _i
,a{~l[iJ Ulf!J. I
.
• ., • •
(b)
1\
[-2] [3]
4
6 . ' -9
10 15
-6

,,,f!J[IJGJ , ..

Solution:
(a) Here, vi, v2, V3, v, e R3• Since there are 4 vectors but 3 · t . .
~. 1 ;•.tr _',,· , r:·, - '· megers1.e.4>3.
That IS, here number ofvector= -4~ "3 = numL:..
ut:r o vectors. ·
f · ·

Therefore, given set of vectors are linearly depende~t.

M. ~vt=[I1 m~vi~rnF
 Linear Equatlona In Linear Algebra W IC@:# ti
Here,

10
2

3v1 + 2v2 = {~6 ] + {!]-[~]

-9 - o
15 60
~o
Therefore, Vt and v 2 are linearly m
. d ependent.

(c) Let, t

A=[i g i]
Since the set of vectors A contains a zero vector (second column is zero vector). So A is linearly
dependent.

1 EXERCISE 1.4
1. Are the following sets of vectors linearly dependent? H yes, find the relation between them.

00 [ru-ill-GJ ~ ru.ru
(ill)[:].[!] (iv) [!l[il[i]
(v) [ l].[i].[~J (vi) m-[!JHJ
2. Determine if the columns of the matrix form a linearly independent set. Justify each answer.

1 -74 -35 OJ15 .

(ii)
[-4 -5 7
-2

3. Determine for what value of h, is {v1, v2, V3} linearly dependent?

(i) Vt=[-}]. V2 =[ !J. v,=[f]

(ill) v, =[i] ,v2 =[=}]. v,=[ll
" IIATBBIIATICS-11

1. (i) Linearly independent (ii) Linearly independent · (iii) Linearly inde.Pend

(iv) Linearly dependent ent
(v) Linearly dependent (vi) Linearly independent
2 (i) Linearly independent (ii) Linearly dependent
3. (i) Any value of h
(ii) h=6
(iii) Any value of h.

□□□

'·
J l

• • I
 TRANSFORMATIONS -
. '
I I -, .

- t

(' I
I

. '
("
LEARNING OUTCOMES
After stadJJng tbls cllapter you should be able to:
& Introduction to linear Transformations
& The Matrix of a Linear Transformations - ,.-1 l l .,_

a Linear Models in Business, Science and Engineerin~

' I '(
I 2.1
MATIIBMATICS-D

INTRODIJCI'ION TO ~INEAII TRANSFORMATIONS

. essinS· We think
·Toe matrix equation Ax = b is application in computer graphics and signal pr:h fortn we }<noW
5
the object A acts on a vector x by multiplication and produce a new vecto~ Ax: ':1th A that gives the
as the matrix transforms x into vector b. This means every value of x associates W1
value b of.particular Ax. Such association is a transformation.

This concept generalizes the following definiti

iiiiillon or Fundlon
T &om Rn to R.tn
R.n to a \?ectoi,; T(x) in R:

j Example 1: Let A = 0 2
.
be the given matrix and define
T· R 1 ➔ R1 by T(x) =Ax.Find the
.

images under Tofu= [ !3] and v = [:]. (Model, CSIT, TU] .

Solution:
-[2 OJ d th transformation T: R2 ➔ R2 defined by T(x) = Ax.
LetA- 0 2 an e .
Also, let
u =[_;] and v =[:].
Then
T(u) =Au=[~ ~][_;] =[_26] ·
and . ,

··, T(v)=Av=[~ ~][:]=[~]. ;

~
Thus, the ~ges o~ and v under T are [ _26] ~d [~] .
. .'•
Matrix Transformation .J : - •
., '

In above example Tis a transformation that transforms a matrix u(or v) to T(u) (or T(v)}which
,is again a matrix. ! •
.., .\
'This concept develops the idea of matrix transformation.
_<Matti
•-----·••:" on Trandorm~on): For each x E an, T(x) is computed as. . llof•••:' ...
....;./· '. . (

_ mattix behaves as tr~ormatiQn operator. .;,,.,,,;·

. .
. _, - ... - """
1. •.
 1 -3] Tnndmnattona w l§w:r- 21
Example 2: Let A = 3 5 ,u=
-1 7[
T: R2 ➔ R3 by T(x) = Ax
[2] [3]
-1 , b = 2
-5 ,
c -
-
[3]
2
5
. . ~...-,
and define a transformation
(a) Find T(u). so that
(b) Find x in R2 whose ima (2078]
(c) Is there more th ge under T is b. (2078]
(d) D e termine if c isaninone
th x whose lfflage
. under Tis b?
erange ofT.
Solution:

~ A= [!1 -3]~ 'u = [ _;], b =Lt] and c = [~] .

Given that the transform ti 5
Now, a on T: R2 ➔ R3 by T(x) = Ax.

(a) T(u) = Au = [l rJ [_;] !;}] rn =[ =

(b) Let X = [;~ . _

Suppose x in R2 whose image under Tis b. Theri
T(x) = b ⇒ Ax = b

⇒[l ?] [:~ =[!]

~~ au7r :mxir = b ~,

--[l~ . ;! -;7]
0 4 -2 ,
[A I ➔R2
PPY~➔ &+Ri
R2-3R1

~ [~ 7 l.s] [Apply R2 ➔ R2/14 -' .,

0 4 -2

r
-[~ -; -ts]
0 0 0
'
1 0 1.5]
[Apply R1 ➔ R1 -3R2
~[ ~ ~ -~5
This implies x1 =1.5 and x2 =-0.5
Thus, x =[!~]in R2 whose image under Tis b. 1

(c) In above solution (b), x has no free variable, so the solution xis unique. This means there is
exactly one x in R whose image under T is b. ·
(d) From (c), there is exactly one range b of T. So, c is not a range of T. .
- For (d) we can proceed as in (b) with replacing value of b by c. Then we will get an
inconsistent augmented matrix of Ax == c. This implies c is not a range of T.
 MATBBIIATic&-11

Shear Tranaformadon 1

1 3] -[OJ v=[ 2].

,2
f •J

Example 3: Let A = [o t , u - 2 ' ,..T

Then the image of u under A is t,

Au=[~ n[~] =[~]

And, the image of v under A is

Av=[~ n[~] =[~]

T -
~X2 (8, 2)

(2, 2)

0 .
.' . .'
Shear transformati9n -

Example 4: Use a rectangular coordinate syst~m with coordinatt; (?.S, 0) and (0, 0.5) to plot u = [~] ,
- . .
2
v = [ ~ ] and their images under the gi~en transfo~ation T.
Solution:

Let, T=[005 o~s], u=[;]and ~ = [ -;] :

: ..
Then the image of Tis

Tu = [
0
,,5 o~s] [;] 2i5] and
=[ Tv--[0.5
0, 0J[-2]
0.5
-[-1]
4 - 2

.l2)
•!

v(-~

.lfl.
Tv(- 1, 2)
Tu 2.5, 1
 Tnmaformatlona W \§w,#21
JJ!lear 'fr&D&formatlon
A tra115fonn x ➔ Ax has the properties.
A(u + y) =Au+ Av and A(cu) = cA(u)
for any u, v in R 1? and for any scalar c.

fhis concept leads the idea of linearity.

~.u~•ution (linearity of transformations)

A ~furmation (or mappii:l.g) Tis linear if ,
T(u + v) = T(u) + T(v) for any u,' v in domain of T,
T(cu) = cT(u) for any u in domain of T and for any scalar c.
l'he single equivalent condition for linearity of Tis fo;
all a, J} e F, u, v domain of T, is
_ . T(au + {iv)= aT(u) + J}t(v). · ·
Note::IfT ~R0 ➔ Rmis a lineiir'then T(O) ~ 0, ifT(O)*O then Tis not linear.

Example S: Prove that contraction and dilation map is linear transformation.

Proof: We know that map T : R 2 ➔ R 2 defined by T(x) = rx, where O~ r ~ 1_is called con~action map.
Let u, v e R2 and c and dare scalar. .. ., ..

T(cu + dv) = r(cu + dv)

'"- ,,_.
= rcu + rdv
,. , =·~(~) + -d (~) ~-, j:
.. .. r I
r1 ..:

= cT(u) + dT(v)
This means the map T ~pp.ear. .-. -< ~i .,ru · u , )
Also, the map T ~ lli)ear whep T is dilation. T • ' I - r -

&ample 6: A mappin~ T: R~R2 is,4efined by 'f(x) =[~ -:J[:] =[-x71- tu. Find Tv and T(u +

v)whereu=m~~vs[!J. .. . . . .. [20'7&]

Solution: Let
'

Then

Therefore,
 ·- 'T'
MATHBMATICS-ll

BXERCISE 2.1 ,, -
2
A
l
1
. . derTofu=[- ]andv==Fal d
1. (i) Let A= [ ~ ~] and define T: R4R2 by T(x) = Ax. Find lhe unage un -; Lbi·
(ii) Let A = [ O 1] and define T: R4R2 by T(x) = Ax. Find the image under Tofu =
-1 0
and v
[2078]
[-3] =[s}

(iii) Let A =
[05 ~
0
0.5 }Ju= [!]and v = [
0
n Define T: R'➔R• by T(x) = Ax. Find T(u) and T(v).
2. (i) LetA= ~2 [1 1
-2
0
-:] , b =
-5 -3
[i].
Define T by T(x) = Ax. Find a vector x whose image under T

is b.
(ii) Let A=[_~ -5
~1, =[ =;J.
b De~e T by ! .(x) = Ax. Find a vector x whose image under T

;::e
7

e[r t r1p]ped
is b.

3. : :: ~ :e
v:::. ttamformation x➔Ax for the given
2 -6 6 -4 - I'

4. Let b
0
= r~lJ and A =[~
2-6
i. : -:] .Is b in the range of linear transformation x➔Ax?
6 -4 .
Why

or why not?

5. Use a ~gular coordinate system to plot u = [~], v =. [ i] and their Images under the given
transformation T. Also; describe geometrically what TUoes to each vector x in R2•

(i) T,(~) = [ ~l :.~ ] [;~ T(x) F ~~ -~] [:~ • . . .. ~,

.
- 1. ..

1. (i{=~J, [:]. .-·(iii) [

0.5 ] [0.5 a]
0 , :Q.5 b
... -2.0 0.5 C
2.(i) rn .
~fr]
I

3. X ~ ~[i]+ 4. Yes, because system represented by[~ b] is consistent I ·• i1

5. (i) (ii) X

V•- T (v l
u
------u
-t---,t--+--+---+--l--+..:..+-4---1----+.-x ,

•rrv)
A reflection thr ugh the origin
A projection ont~ the xi-axis
 Tranaformat1ona w ICii§K #I
,.~,. TUE MATRIX OF A LINEAR TilANSFORMATIONS ·
- }ineat transformation T:RrL+Rm is defined as T(x). Which is actually a matrix a transform x ➔ Ax is
th
~esr that means has e properties of linearity. The key to finding A is to observe Tis completely
rauned by what it does to the columns of nxn identity matrix 1
dete . A
Theorem I: Let T: Rn ➔ Rm be a linear transformation. Then there exists a unique ma
such that T(x) = Ax for all x in Jln. In fact, A fs mMn matrix whose jth column is the vector T(ei)
th
where ei is the j column of the identity matrix in Rn: A= (T(e1) •.•••• T(en)],
rroof: Let

Let Tis linear. So,

T(x) = T(x1ei + X2e2 + ... + Xnen)
.. r ~ i1 T(ei) + X2 T(e2) + . .' . + Xn T(en)

A [T( )
=Ax
T(o..;)
= ff(e,)

T(e )]
T(e:z) . . .

is called standard
T(e.)l [:~

matrix for
'
linear transformation T.
I
.I

where = et "'" · · · n - f all in

'bl
If poss1 e sup pose that A is. not unique. Then there is a matrix B
, __ such that T(x) - Bx or x
Rn.
Therefore,
Ax=Bx ¥ XE Rn
⇒ {A - B) X =0 ¥ X E Rn
Let A - B = C be mxn matrix
So, Cx=O

Let we choose,

Then
C C12
c:: C22
Cx = O or Cei = 0 ⇒
[
.
Cmt Cm2
4 .
. ⇒ Cjt -
- 0 1· =. 1, 2, ...., m
¥
.
. d . _ 2 m. Then,
s:-:1arly c·· = 0 ¥ i = 1, 2, ...., n an J - 1, ' •.•- , 0
UJW ' lJ • A - B-
C = 0 i.e. I .I

⇒ A=B
This means the matrix A is unique.
 MATBEMATl~U

• D2
3x for x Jl\ .I'- •
Example 1: Find the standard matrix A for linear transformation T(x) =
Solution:
Let T(x) = 3x. In R2, . O [OJ
T(e,) = 3e, = 3 [~] = [~] and ~
T(e,) 3ez = 3 [1] = 3
Now, the standard matrix A for T(x) = 3x is,
A= [T(ei) · T(ei)] = [~ ~] ·

3
Example 2: Find the standard matrix A for lin~ar transformation T(x) = 2x for x in R •

Solution:
Let T(x) = 2.x. In R3,

T(e,) = 2e, = {~] = m

.
I,

T(~) ~2e,={i] =m

and T(e,) = ~ = {~] = ml

Now, the standard matrix A for T(x) = 2x is,

A = [f(e,) T(e,) T(e,)J = rn ! ~l

Alternatively,
Given T{x) = 2x
i.e. T(x~ x2,1x3) = (2x1, 2x:u 2x3)

or, r( ,::) =( ~:) =( :: : ~: ~\ J~

X3 2x3 Ox1 + Ox~+~ )
1
\~
0
2
0

Thus, T(x) = Ax, when A = G ~ ~ is required matrix.

 Tranarormattona W tciwjn21
etrlc linear transformations of a2
c,eoJJI . Jin tr f . .
The geometric ear ans ormatio of R 2 is a plane that is rotated as
e identity. Being the transformation is line , the geometry are
t h ~ completely by what they do lo the colum&..' The images of "1
!d ei shows the transformation gives a Unit.square. • x1

aotatfon Transformation r-
n - 1
,
.
, ., r •.:~
,,
A transformation IS angular form is known as rotatibn transformation. The following example
lores the concept. _ ~ .
exp ' . ..
2
£xaJ11ple 3: Let T: R4R that rotates each Pojnf in R~ with an angle 8, in counterclockwise
direction. Find the standard matrix A idr lineu transformation T in R 2•
LI
Solution: . ~

[1]
. Let ei = (1, 0): 0 and~= (O; 1) = i {OJ . Let the (- sin8, cos8
---...(0, 1)
,..,
vector rotates with an angle 8 -in ~
cos8, sjn8)
countercl~ direction. Then [~] rotates

into [::J and m -=lThat

~lo [ is,

----T(e,) ~ [:!] -- - - ' .,

[ ,,,· 1, ',
and- ----T(ei)-=r·~=:J·: --·- f:[" '::J.
-- -- -
• ·1 • l

Now, the standard mam,c A for Tis/

. tcos8· - sin8] ·, 11 -
A= ff(ei) T(~)] = ·sin8 , cos8 · · .
. Other Trand'ormationa of R2 is Shown in. Fo~q~iig Table;-
. - . .
Table 1 Reflections : ~-_ . ..
Transformation Image of the Unit Sqilat~ Standard Matrix

Reflection through
the X1-axis

J • .• l J

Reflection through
theX2-ilXis ! . >,
,.[-~' o]
0 1
t •.
'......-•.

''
 ----
Reflection through · x2
the line x:z=x1
'ftc
.
}-le
. sh
Xt

Reflection through
the line x2=-x1 • I

;,. . V
sl

~,,., , -·- .
Rcllection through
the.-origin [i ~1] • ,• I

.. tJJ

~ I ..
I , I
( ' 1
\,

Table 2 Contractions and Expansions

Transformation Image of the Unit Square · Standard' Matrix
I

Horizontal
contraction
k_
[ 0 1·
01 1
·. ,
.I, :
and expansion

+-- .
+--

O<k<l k>l

Vertical
contraction
and expansion
1
[0 k .
01

Xt

k>I
 Tranaformationa W ICaAPra#J
Table 3 Shears
~ansfonnation Image of the Unit Square

·-
ttoriZ<>ntal
shear
Standard Matrix

x,
k [~]
. k<O k>O
vector X.2
shear [t ~]
w,
X.1 x,
)

k<O lb O

Table 4 Projections
Transformation Image of the Unit Square Standard Matrix

Projection onto
thex1-axis

!! x, . I

[~] [~J
Projection onto X.2

the x:z-axis ..
_. t~] 4--
. ,:.'
.
... (, . t. . .
.--·· I ) • I

;•
•

1 i,
, I • .,r x, , r d -

[~] .J•

• • I 31t
Example 4: Let T: R4R2 that rotat~s eac~ point} ~ ~ thro~gh ~ angle 2 , in countercloclcwise 2

direction. Find the stancJard.matrix A for tmear transformati~n 't: ~ IV. . (2076}

Solution: , - ( - , 1, T
3 0

· Let ei = (1, O) = [~] and e, = (0, 1) = [~]. Let the ~ector rotates thr_ough an angle ; in
_; J l
counterclockwise direction. Then,
cos(31t/2)]-[ 0] - sin(31t/2)]
and T(~) = [ cos(3n/2) = OJ
[17
T(ei) = [ sin(31t/2) - -1
Now, the standard matrix A for T is,
 IIATIIBIIATICS-11 I,,

rj ..

R' - R' R' R' R' R'

T is one-to-one Tis onto but Tis not onto
and onto not one-to-one

'
Proof: Let T: R 0 ➔ Rm is linear transformation.
Suppose that Tis one-to-one. Then for any x in Rn,
T(x) = 0 = T(O)
l
⇒ x=O [" .. being T is one-to-one]
This means the equation T(x) = 0 has only the trivial solution.
Conversely, suppose that the equation T(x) = 0 has only the trivial solution. And, we wish to
show Tis one-to-one.
Take, T(u) = T(v) for some u, v in Rn
⇒ T(u) - T(v) = 0.
• •a"') •' 1 I • •' •

⇒ T(u - _v) = 0,
being Tis linear. . Jl

Since T(x) = 0 has only trivial solution. So, ~e ~hould have,

T(u-v) = O I

⇒ u-v =O.
⇒ u=v. I t t

Thus, T(u) = T(v) ⇒ u.=v.

This means Tis one-to-one.
I
 r,oo{: Let T: R.~ R.m be a linear transformation and let A be the standard matrix for T.
I

(a) Let
Tis onto ~ for each b e Rm 3 x e Rn such that T(x) = b
~ for each b e Rm Ax -- b h as solution,
. w h ere A •is. mxn matrix•
I •

~ column of A span Rm.

(b) Let
T is one to one · ~ :,equation T(x) = Ofhas only the trlvial solution.
., , • .,J ' ~

~ Equation Ax = O·has only trivial ~lution

~ Column of A are linearly independent.
. • r , 1. • , T
example S: (i) Check the transformation T defined by T(x, y) = (x, 2y) is linear or not. I
• • j

(Model, CSIT, :rul

(ii) Is the ~ansformation T:~2i+R2 d~ined ~y T(x, y) = (3x +
' • '
1
.,. •
y, Sx + 7y, x + 3y) linear?
{2075}
(iii) Show that the transformation T defined by T(xi, x2) = (2x1 - 3x:z, x + 4, Sx2) is not
linear. · , · ·- '' · ·
Solution:
i I I
(i) Let T is a transformation, defined .b y
') .... 11 'I T(x, y) = (x, 2y) · f" + '· ·'
r,
Set U = (x1, x2) and V = (y1, y2)" ......._, . : ., ~ 1

Now,
T(~ + v) =T(x1 + y1, x2 + y2) =(x1 + y1, 2(x2 + y2))
i = (x1, 42), + (y1, 2y2) I

I I
,h to = T{x1, x2) + T(y1, y2)

=T~+Tv
) .
This implies that Tis a linear transformation.
! '
(ii) Le\ ;r is ~ transfo~tiqn, defined by I •t

T(x, y) = (3x + y, Sx + 7y, x + 3!! . ,,

,.
Set ·u = (x1, x2) and v =: (y1, Y2)
t • , I
• I "~

Now, . .
T(u + v) = T(:>11 + Yu x2 + Y )
2
+ ,; : (x + y2), + y,) + 3(X2 + y2))
7 2
(x,
= (3(x1 + y1) + (x2 .+ y2), S(x1 y 7 + y ))
) + (3y1 + y2, 5y1 + y2, y1 2
= (3Xt + X2, ·5x1 +7x2, x1 + 3x2
 THEMATICS-II

and, 2) S + 7y2, yt + y2))

T(u) + T(v) = T(x1, x2) + T(y1, y + 3x2) + (3yt + y2, Yt
= (3x1 + x2, 5x1 +7x2, X1 • • • ! .

= T(u, v).
. .
This implies that TIS a linear transformation.
(iii) Let T is a transformation, defined by 5

T(x1, x2) = (2x1 - 3x2, x1 + 4, 5x2). . -. ~

Now, ) 3(u2 + v2), (ut + Vt· ) + 4' 5(u2 + )v2))

T(u + v) = T(u1 + vi, u2 + v2) = (2(u1 + v1 - . v ·Ut + Vt + , 5 + 5v2
4 02
= (2u1 + 2v1 - 3u2 - 3 •2,r .:
and,
T(u) + T(v) = T(u1, u2) + T(v1, v2) + Sv )
I
. = (2u1 - 3u2,~n1 + 4, Sui) + (~Yt - Vt 4' 2
3v2,r- j I;

• - "II' ' 5 )
"'.' (2u1 + 2v1 - 3~r- 3v2, Q.1 j- v1 ,+,B,15~2-: V.2
T(u, Jv) . . _ , -, . t..
:;i! •• , .m . '.) • ·•
,1 1

This implies that Tis not a linear tr~fo~tion.

Alternatively,
• j •
. ltr, - :>rH ; :,
) ' )

For tfiis transformation f . • • , ~. _ ~• "

T(x1, x~) = (2xt ~ 3x2, Xt + 4, Sx2)
1

'
r- • I , • ., •-

Tak~ (xi, x2) = (0, 0) then__ . · .. • , . ., ·, · ,. ·; , ·

•T • ' *
T(0, 0) = (0, 4, 0) (0, 0, 0)
⇒ T(0):;!0
This shows T is not linear.
Example 6: Let T(x1, x2) = (3x1 + X2, Sx1 + 7x2, X1 + 3x2). • . (Model, CSIT, TU]
(i) Show that T is a one-to-one linear transformatio~.
(ii) Does T map: R 2 onto R 3?
Solution: .
I. T .•

Let,

Then
3x1 +x2]
T(x) =[ 5x1 + 7x2 =5
[3 ~][xil_= Ax (let). -
X1 + 3X2 1 xJ ..
3 .
(i) Since we have (by theorem 3), T is one-to-one linear transformation ·if and only if
columns of A are linearly independent
Here, A is 3x2 matrix in which one column is not a multiple of another. This means*
columns of A are linearly independent. Therefore T is one-to-one linear transformation.
(ii)
Since we have (by theorem 3) Tis onto if and only if the columns of A span R3•
Oearly A has
3 only 2 col~. So, it has at most two pivot positiOIIS. This meaJ1S A
not span R • Therefore, TIS not onto. .
 ~aJllpJe 7: Let T:. R4R4 be a Jin 'rranaronnatto §APn:R
is onto mapping. Justuy~ar transformation h na W ~ ?I
, [ Your illlslV w ose standard
-5
8 10
3 -5 4] er. , Dlatrix is given. Describe if.T
~ ' . -4 7
4 -9 5 -3 . '
-3 -2 5 4
1
s~Jutiofl: I r

[:
10
3
Here,
-9
-3 -2
10

[f
-5
95 60
-5 -5
-40 40
-5
0 -

i
[
~ .r

I o r

• f - I ri •

ruJ
This shows that ea~ ' ~oluinn has~pivot'~lement, Tis onto. -
~ote: The above example·shows the column vectors ~e linearly independent, so the mapping T
JS one to one. - ~ 1
• I I

..1.,EXERCISE 2.2
1. Assume that Tis a linear transformation. Find the s~dard matrix 'Of Tl
(i) T: R4R4, T(e;) = (3, 1, 3, 1) and T(e2) = (-5, 2, 0, 0) where ei = (1, 0) and ei = (0, 1).
(ii) T: R3➔R2, T(ei) = (1, 3), T(e2) = (4, -7) and T(eJ) = (-5, 4) where ei, eu e:J are columns of 3x3
identity matrix. _ .
2. Let T be a linear transformation· whose starldilrd matrix is given.VOescribe ·if T is' a one-to-one
lllapping. Justify your ~er..

(i)
-5
8
10
3
-5
-4
4
7
\
(ii}
I [ 7
!~ :
5 1~
12
!]
7 ·
\ .
I' J ,

[4 ~ 5 ~ 5
-3 -2 5 4 -8 -6 -2
 . MATHEMATICS-ti·
~ . · . ·Jin af also find the matrix that implemE!nts the ll\a
3. Prove the following transformation T is : X4) . . • ~t
(i) T(xu x2, "3, X4) = (0, x1 + x2, X2 + "3, X3 - • ,
6
(n") T(x1, x2, X3) = (x1 - Sx2 + 4"3, x2 - . "3) h th t T(x1 x2) = (x1 + X2, 4.xt + 5x2) . p·md x su h
Let T: R4R2 be a linear transformation sue a _ c ¾t
4. , ·

T(x) = (3, 8). · d d atrix is gjven Describe if T ·

5. Let T: RS ➔ RS be a linear transformation whose stan ar m . . is onto
mapping. Justify your answer.
4 -7 3 7 5
· 6 -8 5 12 -8
(i) -7 10 -8 -9 14
·3 -5 4 2 -6
-5 6 -6 -7 3 j

9 13 5 6 -1
14 15 -7 -6 4
(ii) -8 -9 12 -5' -9 :,
-5 -6 -8 9 ' 8
13 14 15 2 11
·_. ~: (i) · ff linear transformation T: R4 ➔ R4 is definE:d by T(x1, xi,·::xa, ~) = (0, x~ + x2, x2 + Xa, xa + '4).
J ··- Check it is (a) one to one (b) onto. '
(ii) ·ff linear transformation T: R3 ➔ R2 is defined by 'r(x1, '-x2, xa) = (x1 - · Sx2 + 4:xa, x2 - 6Xa).
Check it is (a) one to one (b) onto. _ :'
• - ,,. - t
7. Let T: R4R be a linear transformation.. Find the standard matrix:1\ for linear transformation T.
2
i, .-

'
J , :-:
I - -1t ' - j

(i) T rotates each.point in R2 through -a n angle 4 ,~~ cloc~se directiQn.'

· -371l ii -
(ii) T rotates each point in R2_~pu~ angle ~" , in ~l,< ;>C~ dir~on and then reflects
the point through horizontal ,k axfs.
1 i rj,5 ·c·· •J ,1
IL'.,, • r
I

L (i)[ifl . ;2. .(.i) Not (ii) Not

', '

J. (i{f iJ!J T~:iJ ~il) [ ~ . ,. :J' •

5. (i)'Not ' (iif-' Not · '· .,

6. . (i) No~one to.o~e, not onto -(ii) Not one to one, but onto \'

7. (i)[ 1/V2
-11vi (ii) [-1/-{2 11V21
-1/-{2 1/~2.l
 ,.3 LINEAR MODELS IN BUSJNE Tranarormattona W f§w;;aj
SS, SaENQ AND ENG , . t •:--
liJtear model of equation has Wid . INEERING
fhe . Th lin . e apPlication . diff .
-•-ost linear. .e
are ilJ.U•. ear ..u"---'tial equation ism
• ~iicren
erent field . Often, the natural phenomena
eeonoaucs and telecommurucatton. powerful tool for engineenng and equally in

1,111ear Programming Proble..

. .
~pie 1: If possible, find some combh..... ti · '
A&.. on of nonfat ·nc, · ' ·
d f mt s~~ fl~~, and ,w h1;r to provide the
-- exact amounts of protein., carbohydrate
Amounts (g) Supplied per lOO of
g Ingredient
' ~ at supplied by the diet in on e day.
· ·
-- Nutrient ·Nonfat milk s fl Amounts (g) Supplied by
oy our Whey Cambridge Diet in One Day
Protein 36 St ' .'.
Carbohydrate ~~
""" . ,
34 'r -"
·-,4 ,1 1 • , .., , ., IE.
"45 ., .
Fat 0 7 1.1, , • I ,
3
Solution:
Let xi, x2 and XJ, res~vely, d~ote ~e number of ~ ~·(loo ;· ) of these foodstuffs. Then
I 1 • l · ,1-~ ' ., •l . l g
., x1 + 51x2 + 13"3 ="33- i ,
1
~•••• (ij 1•

52x1 + 34x2 + 74XJ = 43 •••••(ii) n J/

1

• . Ox1 + 7X2 + 1.1"3 = 3 ·,C:J .,t!_.(ifi) ,' • ; • Ul ' !

Row reducti_on o~ the a~~ented matrix for_the cov~J>8n9ing_system of equations shows that
·,[3652 51 13 33]
34 ' 74 45 ~
'[ 1 . O
~ 0 1 0 0 392
0 0.2ill l' ' , .-
ts
0 7 , Ll , 3 ., _· .. ~. . 0 - 0 . l 0:233 •
_t.. '. I · . •·J, :,_ - - .I- -c.: ·'I ·- g· 10-l"lJ !6 ,~"' , •t • n· ..,, _
To three significant digits, the diet requires 0.277 units of nonfat milk, 0.392 units of soy flour, and
0.233 units of whey in order to-provide the desired
'
amouhts
of protein, carbohydrate, and fat.
.
lt .,. ,
Engineering Problem
•; • , ,, , . • ( •t.1 I ·; , -,, fI
Kirchhoff's Voltage Law ·· · ' · ·-:;' ~, ,. ~ J - • •

The algebraic sum of the RI voltage drops in one direction,~qµnd aJ~p ~u~ the algebraic
sum of the voltage sources in the same direction _a roun~ the ~oop.

Example 2: Determine the loop currents in the network in Figure.

• ' I I
· ~ ··,..;.
r;...;.:,
.,.J
· ,-It----,

.... '

20 vo115
 MATHEMATICS-D
I
Solution:
In loop 1: .
. 30V Total RI voltage drop is • I
Current l1 and h flow and voltage is ·
411 + 3lt + 411 - 3h = 111t - 3h
BA) .
(Here in loop 1, h flows opposite to It along
I

Thus by ~choff voltage law

I .._ .. 11Ii-3h =30 ····· (i)

In loop 2:
.
Current l1, l2 and .h flows and voltage is SV. Total RI voltage drop is

- 311 + h + l2 + h + 3h - h
J
= -311 + 6h- h
~ I f I j - I _, • • • • • ~ f • • .,

J (Here in loop 2, 11 flows opposite to h along AB, and h flows opposite to h

along DC) I
'P'tus by .Kirchoff voltage law

' ,..

In loop 3:
.

Current Ii and h fl~ws and voltage is - 5 ..:. 20 = - 25. T~tal RI voltage drops is
: C
'.: .... ("·
,.
<.
11) ' l l

~ h ! h+¼+h

(Here in loop 3, h flows opposite to h along CD)

Thus oy Kirchoff voltage law · . l I I

'
L
.. ~. ! I.i..: ,. '1 ,

.. .. (iii)
I f" i;
Hence, loop currents l1, l2, h are found by solving system (i), (ii) and (iii), which are
1111 - 3h + Oh = 30

- 311 +.6Ii - h = 5

Ol1 - ~2 + 313 = - 25
Augmented matrix is

-5/3
-25
J
-1320/3
 from third row, Tranaro~ttona w t§w:# 21
169 1320
3.h =- 3 ⇒ IJ =7.81 amp
from equation (iii), ·

-h + 31J = -25 ⇒
- Ji + 3 X 7.81 =_25
⇒
- 12 = -25 + 23.43
⇒ l2 = 1.57 amp.
From equation (i)

Jll1 - 312 =30 ⇒

1111 - 3 X 1.57 =30
.. , ⇒ 1111 =34,71

⇒ l1 = 3.15

Thus, l1 = 3.15 ~p, l2 = ~.57 ~~ ~~ h_= - 7.si amp.

. ' .
J '
Dynamical System Problem
Difference Equation

If Xk be the state of the system at the time of k th measurement and if there is a matrix A such that
Xk +1 = Axk . Cf. .. . (i)
then (i) is called a liner difference equation or recurrence relation.

Example 3: Compute the population of a region for the year 2001 and 2002, given that the
population in 2000 was 6,00,000 in the city and 4,00,000 in the suburbs, whose
. . mab'ix.IS
mt21"ation
Migrate from
City Suburds I"--- To

0.95 0.03 City

0.05 0.97 Suburds

Solution:
Given that, the initial population in 2000 is
6,00,000]
Xo = [ 4,00,000

And, the migration matrix is

0.95 0.03 ]
M= [ 0.05 0.97
We know the linear difference equation is
Xk+l =MXk
for 2001
i.e. Xt = MXo
for 2002.
and, X2 = Mxt
 MATHEMATICS-II

For 2001,
0.95 0.03 ] [ 6,00,000 ]
Xt =Mxo =[ 0.05 0.97 4,00,000

0.95 X 6,00,000 + 0.03 X 4,00,000 ]

= [ 0.05 X 6,00,000 + 0.97 X 4,00,000

-[ 582000]
- 418000

For 2002,

0.95 0.03 ] [ 5,82,000 ]

X2 =Mxt =[ 0.05 0.97 4,18,000

-[ 0.95 X 5,82,QOO + 0.03 X 4,18,0QO ]

- 0.05 X 5,82, 000 + 0.97 X 4,18,{)()Q

- [ 5,65,440 1
- 4,34,560 J
l I
.....
Thus, in 2002, the population of city will be 5,65,440 and of the suburbs will be 4,34,560.

Iii
□□□
_., . .. ,.. .. , I:, J

•J
 I I
' ;

·MATRIX .ALGEBRA 1

I · 0 ' 1 I

e t .io 1

I t ,.; n·_·,' -
4 ..
J, ; ; , • . I r,.l!r. ':nH fI .... 1 .1 £1. - '(_fl:., :I ',

After studying tbls cllapter you uoul be able to:

.J

& Matrix Operations

'& The Inverse of a Matrix
'& Characterizations of Invertibl~ Matrices .
& Partitioned Matrices . J J I I L
& Matrix Factorizations
a. The Leontief Input Outpu~ Model . i

a. Applioatio~_of ~trix Algeb,a to Computer Graphics _

a. Subspaces of 9t1
, IIATBEMATIC8-D

INTRODIJCTION Mat
If A is mxn matrix .in which A has m rows and n-columns. For instance, the x3 matrix can~
written as,
2
' mat

2 1 6]
· . A= [ 3 2 4
The entry aq be the (i, j)lh entry of ~e matrix. If A is mxn matrix then we write it in matrix form as,
No
. [at _ __at ____ ajt --+row sec
A• an ____ aij ---· a1n
.. .. ..
. . .
am1 ____ amj ____ a So
i
column
Thus the above matrix can be written as,
A• [a1 il2
r:
• • • . an). -· ,,,
·
•t
J
'1,

We already mentioned tfiat aii be any . try of I/{. The ~ ·es aii1with i ':= j, iY~ed leaping diagonal

I
entries of A. • · , - · · · · ' ·· ·

3.1 MATRIX OPERATIONS

Sums and Scalar Multiples

HA and B are two matrices of same size (same size means they have same number of rows and
columns). Then the sum of these matrices be the sum of corresponding entries of the matri~es.

HA be a matrix and r be any scalar. Then the multiple rA is the multiple of each entries of A.
Example 1: 1 iet,
2
A= [ 5
3
6
4]
7 'B = 2
[1 -56 •· 1]3 --~

A
+ -[2+ t ..
B- 5+2 6+6
s-s · ij~~11 ~[3•~-,2. ·, s IJ7 ·
7+3J - 7, 12 10
Also, Jet r = 2. Then,

. ,. 1
Note that the subtraction of matrices A and B be redefined as, ,·.
A-B=A+(-l)B.
. The following theorem shows the operations of matrices and sealars.
· ·
Th.w.,.... 1: lit ~ B ancl C be tfi.a,. • es·,o .the same size, let r and s :--
be-:·s-cal
-=ars
- · _,....,~~11,,l,f'":'-"ifl~""--rar"':
(a A +13 = B +A ~ (b)· (A+ B) + C=A+ (B + q
c A +0 = A (d) r(A + B) = ~A + rB .
rsA = rs A
 ,trlx Multiplication Matr1a Ai-
,. 1
JT• •-.soabra W l§iw:# !I
If A is mxn and B is nxp mat-=~- r
d ~~-~th ·
Jllatrix of or er mxp whose columns e col\Ul\ns of B
are Ab1, Ab are, ht, bi, ... , b Th
AB = Afbt 2, • •• , Abp that is, P· en the product AB is
bi . . . hp]
'
>, = [Abi A~
· · · · Ab] · 1
Note that: For the multiplicatio th P • ,
n, e numbe
second matrix, should be equal r of cohunns of first
· matrix and the number of rows of
example 2: Compute AB where A = [2 3] . [
1 -s ande. 4 3 6]
Solution: 1 -2 3 -•
Here, the columns of B are,
bi=[14J, ~ = -2 [3]" and ~=[;].
Then,

=[f ] (4] == [ 2x4 tJxl ] [ll]

3
_Abt
,_ , - 5:_ } , , 1~4 + (-S)xl ~- _1 .

-- A~=[i -53][3]=[
Vt..

2x3+3x(-2) -] [OJ '"'

-2 ' 1x3 + (-S)x(-2) 13 . =
- '
, I . • •t

Al>J =[i -53][6]-[

.. f'i h) Jr I J ..

md
2x6 + 3x3 ] [21]
3 - lx6 + (-5)x3 -9 . •, =
' J
., I

Now, ,Jf'l1 /{j

' AB=Afbi
= [Abi ' I

=[11-1 130 -921] · .,

Another method:

• J • __. I•

=[11 0
-1 13
Deffnltion (Transpose ofa Mcltrix) -
If A is m><n matrix then th~ transpose of the given matrix A is denoted by AT 81\iit'Hat isnxm
lllatrix. Thus, the change of row to column and column to rol_V-' of entries of a given matrix A,
is the transpose of A
 MATHEMATICS-II
I t
2 54 61] . Compute AT.
Example 3: Let A = [ 3 •11I fT :
. , ... So
. fl
Solution: ' ., I
Here,

A=[; : ~].

Th~ AT~[~ :J. . · (

(
Theorem 2: Let A be an ~ xn matrix, and let Band c;_l}ave sizes for which,the indicated
and products are defined. tt'.1,u~u-i:!ir.:+1

(a) A(Bq = (AB)C (~ssociative law of ~ultiplicati?n) ·: r \ • ,

(b) A(B + q =AB+ AC (left disb'fbuti~fl~w)' ·• . I'
(c) (B + C)A =BA+ CA (righ,t distribu,tiye iiW},,
~ t ., <:~/·;_,-.,, :,,,<: :\_ ·.,_-- ..>'
(d) r(AB) = (rA)B = A(rB) , , for apy scalar fa:w:.-1t,, ;';: .. ,
,( ·', ' -· ,,"/1-- ~-7· -\· ·-;

(e) ImA =A= Ain (idehtity~for,1ma~\~~ #J>:lic~tion} .

Proof: Let A be an mxn matrix and Band C ~~e ~izes f~i which the indicated sums and products
are defined. . .
(a) Let the matrix C can be written with 'its columns c1,-c2, ..., cp; as,
C = [c1 c2 . . . cp]
· (- _ t_ -

Then by multiplication of matrices,

BC = [Bc1 Bc2 . . . Bcp]
"
, , I

and,
A(BC) = [A(Bc1) A(Bc2) ....... (i)'
Since by definition
A(Bx) = (AB)x for all x.
. ">· l.. -;
So, (i) becomes,
A(BC) = [(AB)c1 (AB)c2 (AB)cp]
= (AB) (c1 c2 . . . cp]
= (A~)C.

(b) Let the matrix Band C can be written with its columns as in the form.,
B = [bi ~ bn]
and Cn],
• I' • •
The~ ·
 Matrta Algebra Ill e§# 31
A(B + C) = [A(bi + C1)
= [Abi + Ac1
= [Abi A~
=A[bi ~

=AB+AC.
(c) similar as (b).

(d) Let the matrix B can be written with its columns .

B = n..... 1-.._ as m the form,
LVI U2 ••• bn]
Then by multiplication of matrices,
AB = [Abi A~. . . . Abn].
Let r be a scalar such that
rB = [rbi r~ ... rbn] .
and r(AB) = [rAbi rAbi rAbn]. '
= [Arbi Ar~ ~ ,. Arb~]. ~
ucts · = A[rbi rbi ·· •. rbn].
"
= A(r[bi bi bn]) .
.I
= A(rB) .. -
. '.>b . -
Similarly, r(AB) = (rA)B.
Thus,
. · r(AB) = (rA)B·= A(rl3). fH l •

(e) Let the columns of matrix A are

Let Im be an identity matrix. Then lmX = x ~or x_.~ Rm. • ,
Now, r ,, 1-; y

ImA = Iin [at . a2 ....

.. an]. . ,,.,.#
= [Imai Ima2 ... . l~tt1• I I

= [a1 a2 an],
=A.
" .>

Similarly Ain =A [ei ~

~
en),
'I 'Ji
.
...• •, Aen],
I

= [Aei .A~ , "• :

= [a1 a2 ... an], - • 1 .,,

=A
I; .It
Thus,
I •1 i
Al ' I
Note: If ImA = A = n• • AB and BA are possible. But AB [email protected] not equal to BA.
A and Bare square matrix of same siz.e then
 MATHEMATICS-II

Example 4: Verify AB* BA when A=[~ :J and B = [; ~] · -

(d) Lei
Solution: Th
Here, n,
OJ =[14-2
3
Al
A
and
Tl
T
Thus, AB :1: BA. ~

Th orem · 3: Let .fA ·an 13 denqte matrices 'wh~se siz,e_s are appropriate for theJ ollowii:tg
.. , "· ~

prp<!u~. ,, . , . ' T +-BT

) ~.(AT)T;::: '>A ••
c) .F9r an §C~arr, (iA)
Proof: Let A = (a11 )m•n
(a) The (i, j) th element of A = G, i) th elements of AT
= (i, j)th element of (~1)T

Now,

(b) Let, A= (aij)m•n and B = (bij)mxn are two matrices of same order. So, A+ Bis defined. Therefore
(A+B)T is of order nxm. · ~ ,.
Again,
'AT = (aji)n•m and
Now,
the (i, j) th element of (A + B)T
= the 0, i) th element of (A + B) I r. t \

= the O, i) th element of A + the O, i) th element of B

= the (i, j) th element of AT -f the (i, j) th element 'of BT
= the (i, j) th element of (!AT + B1) -,,
Therefore, (A+ B)T =AT+ BT

(c) Let, A = (ai;)m•n be a matrix. Then

The (i, j) th element of (rA)T = the O, i) th element of [(rA)T)T
= the O, i) th element of (rA)
= r the 0, i) th element of A
= r the (i, j)th element of AT
= the (i, j)th element of r AT
.. (rA)T= rAT
•
 d) Let, A = (ati) be a matrix of order tn>< llatrla 1
(
'J1,ell AB is defined and its order is n Olld let.B .. Cl>i1) be Alaol,,i111 5.,
- •
,t,
-
Therefore (AB)T will be o f order >< tn><p. - a ntat:fix-of order n><p.

Again, let AT = (cli). be a ma . p tn.

trix of ord ·
A)so, let BT c (dq) be a ma . et n•Dt •uch .. ' •
trix of order that clj ,., a
Therefore, BTAT is define d and its or p><n . ,such th at dlj ,., b;i, JI·
'fhus, (AB)T and BTAT .
· -~ 0
£sameor,1 d~ 1S p><m· ·
Now, ~er.

The (i, j)th element ·o f (AB)T

=the,.v, 1·)th element of AB.
= clj1 bu + a. 12~ + • . . + a1nbru
n
=L ajk bia
k=l
n
=L Ckj dik
k=l ,,.
n
L d& Ckj -= the· (i'J·)th element of BTAT.
= k=l
refore
1
Thus, (AB)T = BTAT. . V 1

3.2 TIO IiffEIISE OF A MA~ :·:-.-~:, ''': .I .. . . .

....IX !11, .... ,. • I
r
MATBBIIATIC&-11
;c
,., , t 1J1 ' , ( r)

Example 5: Let A• [ ! !1] and~ • [ t -SJ

2 • ~
solll
Solution: 1

Then,
Ac•[! !1lJ -n ~rn ~J= 1··· ·,,
and
7
_
CA • [3
-sJ[2 ·s].=[1
2 -3 -7
o]=i
0 1
. /. ,

. · erse miitrix of A ,
Thus, AC = I = CA Therefore, A is invertible and C 19 a mv ·

Theorem4:LetA•[: ~.lfad-bc 0
1
A-t • ad - be -c
[d -bl
aj·
If ad-bc • OthenAisnot inv

Pn>of: Since A = [: !J.

Let Ax = 0. H A-1 exists, then
A-1Ax = A-10 ⇒ Ix= 0 ., ..
⇒ X =O. ►•
Therefore, for existence of A- , Ax= ·o has trivial solution.
1

Here, the augmented matrix is, ,.

.,[A OJ =[: ~ ~] 'i; H t.' n.9rn ·, i ' '..i ,1 1 ~•11
' .
1 b/a Ol [1 b/a OJ
~[ c
d oJ ~ 0 ad-be/ a 0 l /. • ~ • I. : ) '- • rf}'

is ech~Ion form and Ax = 0 has tri~ solution on!¥ _wher:i-ttte:r .1 ~ no one free variable1
- _ _ad.-= be n·:-
.d. - ,__: n - - •~-~- :;:...:;__;_.- -~-
1;e. a ~ - ⇒ a -- [}\;~ v; - - - - - · ·- -
- -
-- - - • - - - - - . -
- : _ --

Therefore, for existence of A-1, Ax.= 0 has trivial solution

i.e. ad- be ~ 0. •. I,
' I I •·

Hence, if ad - be~ 0 then A-1 exists.

. · 1 (d -b~ (a b'\ (1 O'\
Here, A-1A =ad_ be -c a) \c _d) = \0 1) = I.
I•! I

Also, AA-.t = I.

Thus A is ~vertible and A-1 = ad _ be -c a .

1 (d -b}
H ad - be = 0, then
1 (d -ab~
A-t = ad - be -c )
which is nQt exist being ad- be =·O., /... ,I . r

This means A is not invertible whenad - be = 0.

. .1'.Jote: The matrix (!.7)is,adi?irit '118~ ~f (~ ~ -~ •~d _;be is d~~ ~f the roal!P
. ' .
.-. .
 llatrlxAJ&eln W \§wjsaj
~pie 6: Find the inverse matrix of A w h ereA= [3
5 !].
solution: '
c.
Given ,.
A [3
- 5

Here, det (A} = \: . !\= 18 - 20 = -2 * 6.

So, A-1 exists. And,

A-l = Adj (A)

det (A}

=.!.[6
-2 -5
:-4]
3
=[-3
5/2

TMo•m 5: . A is :=n
~vei:tib!e nxn matrix theri f?r-~a~ bin J!--n, the eq_uation Ax= b has the
uni e soluti.on-x. A b. ~~ =s•" . . __ -'.:: : •
Proof: Let A is an invertible n><n matrix. Therefore, inverse of A (i.e. A-1) exists. Take bis in Rn such
• ..,, ~ J ._;, • • : t ) I I ,

that Ax= b. 'I - ~

..

Then we have to show

X = A-1b - t .. (i} . /. ,:' . . . '' .1 ' -

⇒ Ax= A(A-1b} = (AA-1)b =lb= b.

1 1
This shows that x = A-1b is a sohitiorl of ki- =·b. = - ;· ' • :; i

For uniqueness of x, we sup~ y be anbther solution of Ax = b, if poss~ble. Then,

' '
Ay=b
⇒ A-l(Ay) = A-1b
⇒ Iy=A-1b I-
I
I ,
...... (ii)
⇒ y=A-1b
• I I
From (i) and (ii) x = Y• . . j '
£Ax b
~ ·· x is ~ unique solution o
I• ' t
= · • . •
means . _. the'iollowing system '(system of linear equations),
Example 7: Using inverse of a matrtx, solve . . ,
8x1 +Sx2• -9
-7x1 - Sx2 = 11
I -

Solution:
Let the ~stem is in Ax = b where, [_9]
.
A= -7
[8 5]
-5 'x
=[xxJl andb=· 11 :
1
I 'l
• IF

Here, ;,.. ti

8 5 \ =-40+35=-5;,:_0. , , '
det (A) = \ -7 -5
 MATBBJIATlcs-11

So, A-1 exists. And,

A- 1
-
[Adi.Al _
det (A) -
.!.[-
-5 7
5 -5] [ 1
g -7/5 == --
Now, the solution of Ax = b is,
1 1 ] [-9]-[ 2J.
x=A- b = [-7/5 -8/5 11 - -
1 5
. d(A-1)-l=A
. then A-1 is invertible an .
Theorem 6: (a) If A is an invertible matriX , , . AB d the inverse of AB ·
· then so IS an 18
(b) If A and B are n><n invertible matrices, d That is (AB)-1 = B-1A-1.
product of the inverses of A and Bin the reverse or e~. ,£ AT is the transpose
0
(c) H A is an invertible matrix, then so is AT and lhe inverse
A-1. That is, A -1 = A-1 T_
Proof:
(a) Let A is an invertible matrix then A-1 exists. And, 1
AA-1 = A-1A = I.
. ,
This means A-1 is also invertible.

., Therefore, (A-t)-1 =A. • r

f - ' ..... ~ l . ' , . ,, I : j ~
(b) Let A and B are invertible matrices. So, A ~1 and B·1 eXISt. Also,
AA-1 = A-1A = I and BB-1 = B-1B = I.
,1· ,. 'I :
Now, Exa
(AB) (B-1A-1) = A(BB-1)A-1 = AIA-1 :f lAA-1 = I.
and, . ··[ I{' /.... · ··

(B-1A-1) (AB)= B-1(A-1A)B = B·11B :;== Br/Bm·I. · f. ._ I ,.' : t ..

Therefore, B-1A- is,clllinverse..ofAQ. 'ry,"'t is~ ., • ,r.

~ , ,1 J Ul... ., 1~'°i"'\ - _._ _ .A

B-1A-1 = (AB)-1. 1
( - •f(.
(c) Let A is an invertible matrix. So, A-1 exists.
Here,
,I

. and (A-t)T AT= (AA-t)T = :rr = I.

t •

(A1)-t = (A-l)T.
I

This means AT is invertible and an inverse matrix of AT~ (AT)-1. That is,
ff I t I , •,

Examp!~8: . LetA=[: ~an~»-:-.[~ ,;].nenye~(AB):-~ :=B-IA-1. I• . I

Solution:
Here,

det (A)= I:. !l=2*0

and det (B) ·= I~ 21
4 =-2*0
I
.,,
So, A-1 and B-1 exist.
Now,
_ _ _ Adj.B Adj.A.
1 1 .
·' . u -
B A - det (B) det (A) ;:
 llatrla Alpbra w tciw;-31
-.-2]
3 .1[ 4
_d_[4
. 2 -5 -:J _
: -2 . -7
•
-- [-2
= 7/2 -3~J [_s~2 -3J =[-13/2
10 ]
Next, 4 43/4 -33/2 ·
AB=[: !][;
:] =[: ~]
Here,

\
det -{AB) = I: ~I =-4.
So, (AB)-1 exists. And,
(AB)-t = adj (AB)
.. det (AB)
=.l-[ 26
-4 -43 !]
=[-13/2 10 ]
43/4 -33/2 ·
Thus, (AB)-t = B-t A-t.

Theorem 7: An nxn matrix A is invertible if"clllld only if A is row equivctlent to In, and in this case,
an ence of e l ~ row uations.thatrequces ,A to ·1n ~ transform In into A-1 •
An Algorithm for Finding A-1 .
Algorithm: If A is row equivalent to I then [A I) is row equivalent to [I A-1). Otherw~ A-1
does not exist.

Example 9: find the inverse of the matrix

A=[~ ! ~], ., .·
4 . -3 8
by using elementary row reduce augmented matrix.
Solution: Here,

[A
~[o
I] = 14
1
0 13
2 1
6'
o~ o~] ·
-3 8. 0
1 O 3 0 1 OJ [Interchanging first and
~
[
0
4
1
_3 soot ·
2 1 O O second row i.e. R1 H R2]
)
1 O 3 0 1 ~1 [Applying
~ 00 1· I I. 2 1 0 OJ -·~ & ➔ & -4R1]
[ -3 -4 0 -4 1
Q 1
1 0 3 I OJ [Applying
• J
~Opl 21 0 0 &-+&+3R2]
[ O 2 3 -4 l
. 0 0 -9/2 1 7 -3 / 2]
1 0 -2 4 -1 .
~[ ~ ~ 1 3/2 -2 1/2
~ [I A-1].
Thus, A-1 exist,.s and
-9/2 7
4
I -3/').]
-1 .
I ,<
LJ ,

A-t = [ ; / -2 1/2
2 •4
 3

•3We already men tioned that a ma trix A IS . a matrix A-1 such that
• u.....,."b'1e if there IS
. m.-c.1u

Ir (The lnvertfble!~en
be square n><n ma .
LetA a . .true
(a,,\ A JS. an invertible~~' .·th,. ~"
(b) A is row equjvale~~ ~ ~ _e,
'c) A has n pivot positions. \
\'
(d) The equation· Ax
· ..-- .1,
0 has .O•
(e) The columns of a 1f?~
(f) The linear ~(orm~tio~ j
(g) The equalipJiA: = b .'!"" af
(h) The col~ ~~-~~ of 1
~ linear traiisto~~~P:..
There is an nx11.mif;rix:~ So
There is mt·~~ ~atnxJ)
ATisaninv ) . -~~- .... :~- .,._
,:•.., '

,. 'ble linear l ,
A linear ~~:~
SUch that . ,f
S(T(x)) = x and
In SUch case, Sis the ·

n.t!OJ'em 9: Let T: .Rll ➔ >

Then T- is invertible if and

s given by S(x) = A-1x is
.. 1
and

·Pmof, Suppose that Tis invertible. So, T-1 exists.

Since T(Sx) =x for all x in R•, holds where S(x) =A-1x is the unique function'. Then, Tis onto
R•. Fox, if bis in R•and S(b) =x. Therefore, T(x) =T(S(b)) = b.
theorem.
Since, A is a standard Dlalrix fox T ..So, Ar = b. Then A is lllvertible, by inv-ertibJe lllatri,c
Conv..,,,.,ly, And, A is invertible, so, A-• exists. Let 8,(x)
supJJOse
transformation.
I

S(fx)=S(Ax) = =x
A-1 (Ax) 1
= A- x. Then S is linear
T(Sx) =A(Sx) =A(A-1x) =x
This means Tis invertible.
the conditions
· Next, we Wish to show S is unique. If possible, suppose there is another function u that satisfies
U(fx) = x and T(Ux) = x

for all x in R.n.

 Matm Algebra W 1§§ 3I
Then w e ~ to show S = U.
f
·since Tis onto. So, for e ch · ·
a v e Rn. T(x) =
Now, v.
S(v) 'T S(Tx) = x
U(v) = U(Tx) = x
This im' p lies S(v) -- U(v) for each v e Rn.
So,S= U.
This means S is unique.

Example 10: Using the Invertible Matrix

. Theore . /
m, prove that A is invertible where
1
A= [ 3 01 -2]
-2 ,.
.;.5 -1 9
Solution: Here,

.
A =[
-1 9
!S ~ =;]
~ [~
. o -1 -1
~
o .·o -,3 .
1] ~ [.~ .~ i] ·
This shows that A has 3-pivot .. -
invertible matrix. positions. .Theref9re, :b y . invertibl~ matrix, µteorem, A is an

I1 EX~RCJSJf 3. J _·~-
L-----_____:__~ -~ ~ ~~
: ~- ,;,; ru ,,.
~:s:=.....=:- - _ J '
I
.
ltrix for T,
2 0 -lJ [. 3 ~>7 . .
1. Let A =[ 4 1_-5 2- · and B = ~l - 4_] . Then compp-~e (i) ~~ (ii)J ~~-
1
;formafmi

2 Compuie A~51 w~ A_~-[~-4 -; !]. 1 8

l .
..• ll,. ,
..
1

-n
r ' , , ,-> • t .

3. LetA =[.! "t], :JandC•[~ ,

B -. [: Verify that AB= AC. I 1

4. Examine matrices are sik~ or non-sirtglilar.. l -~~

1,Tis ~
(ii)[; ~] [2 0 -1]
(i) [~ ;] (iii) 5 ,1 0 -

-n .
0 1 3
· [1 -[]
C: J '
'
ble a,attit 0
-2

F fl
8 1
1 (v) -1 · 5 (ri)u . .
(iv) ~ 5 -4 . ,.t, I -3
(

jS JjrP 0
5. Find the inverse of the matrices by elementary.row red:"lc~ augmented matrix. If exist l

(ii)[~ ~] [2 0 -n
(i) [~ ~J 1,i
(iii) 5
0
1
1
0

(iv>[~
8
(vi[~ -:]
-2
5 (ri) [~
1
-n
1
0 11 -4 -3
 MATBBIIATICS-0 ~·

6. Solve the following system, by using inverse matrix,

(i) 8.r1 + 6x2 = 2 (ii) 3x1 + 4x2 = 3
Sx1 + 4x2 = -1. Sx1 + 6x2 = 7

~~trorufuemamces(:[~v1l•rrmg ve (ill)[! ~ -n
. In rtible Matrix Theorem.

7.
9

1 -3 2]
8. Find a matrix A whose inverse is -3
2
[ 3
-1
-1 .
0

9.
.
Using Invertible Matrix Theorem, show that AT is invertible if A =
[1 0 -2]
3 . 1 -2 .
-5 -1 9

= ( ~l i ) = ( ; !i )
i 10. (i)

(ii)
Let A

Let A =[ _ 3
2
and B

~ J =[ ·: -: ]
B

I
, lJ. Let A =
What value(s) of Kif any will make AB = BA ·

r·; ~ J
Verify that (AB)t = BtN
B=( ~ .~ ] ' · · -- . •
'· l
. ,. '
I
J

26 -25 :...77 -14 , -1

1. AB is not possible. BA = [ 14 _20 7J .2.
[
-8 2
-4 1
4. (i) N.S. (ii) N.S. (iii) N.S. (iv) N.S. (v) S (vi) N.S.

. '
-1 11·
4 3\]
l
-2 ·1 ] 3 '
5. (i) [ 7/2 -3/2 (ii) [ _;~ (~) -15 - .6 -5
, ,5 -2· :. 2
1 -8 .· 31]
(iv) 0
[:
0
1
0
-3J
1
(v) Does not exist
l t
(vi) rl112:o
6. (i) x1 =7 and x2 =-9 (ii) X1 = 5, X2 = -3
.,,
7. (i) Invertible '
· (ii) Invertible
(iii) Non-invertible.

8. [i : !] 10. (i) K = _ 5
(ii) K = 5
 llatrlz Alpbra w 1@#31
:,.4 pAJITl'l'IONED MATBt~s

'Jlte horizontal and vertical dividing rule t .

Difln on (Partitioned or Block Matrix) s o a matrix, such rule is partitions of the m
_ a_t:rtx
_ · ·_. - - -

Ut A be 3X::[T' ~ .'.i ~ ~]
5 3 3 6 3
Divide Aas,
Au=[; ~
A21 = [5 : 3
Then A .can be Written as
A= [AAu AA
· 12 AA~3J . : ~:. . -'·· ·; ... ;, i ~-i -.,,... .
~~ t~~;;~" . . ~~ .
1

21 22~. : 23 ., ~:,,, .. -t?- ~---~~ .. ;

Here, A'is called "tioned or: block matrix whose entries ~e bl~ks A111 Au, A13, A2JJ A-n, A7.3.
Addition and Scalar Multiplication
If two matrices A and B are the same size and are p~tioned in exactly the same way so that
the addition is possible. Then, the sum of A and B is same to the given matrix. As similar, the scalar
multiple of A is the multiplication of a partitioned matrix by a scalar. ·
Multiplication of Partitioned Mahices " .
The multiple of partitioned matrices. by usual row-col~ ·rule, produces a multiple (i.e.
product) of the matrices. · .

The following example illustrates the concept.

Example 1: Let
6 4
-2 1
-3 7 .Find AB
.:f--3
5
I

Solution:
Given
6 4

= 1
2 -3 1 : 0
5 __ :J__._3
-4]
-1_ and B = -3 · 7
1 ____
-2
-----
1
A [
0--~ 2 : 7 -1 -1 3
5

Here
-3 A21 = [O -4 -2],
Au=[f 5
 IIATIIBNATIC&-11
-1
&= [ 5
3]2 . r

· A22 = [7 -1),

Then,

So,

Here,
6 4
AuBt =[i 1
-3][ ]-[15 12]
-2 - 2 1 - 2 -5 .
5
-3 7

A12~=[~ -4] [-1 3]-[-20 -BJ

-1 5 2 - -8 7 ·

'
. ·1 .
J '

Now (i) becomes, ; .

[152 12] + [-20 -8] -5-6
'I

Theorem10(
ismxn
AB =
[[14 _; '. . . -8
-5
r , •_ .
71
~
= •

u!
' I

Example 2: Let A= :J and B = [! •n Find AB by column-row expansion.

l •, f ;1

Solution: ., .
Here,
3 i:
AB = L colk(A) rowk(B). So I
k=l I •

col,(A) row1(B) = [ ;3] (2 1) = [-: · -;J

coh(A) row2(B) = [ 1...] [4 5] = [ 4 . 5 l
.., · -16 -20J

Then,
coh(A) row,(B) = m (6 3) = [!~ lsJ.
 3 llatda Algebra w §-31
Hence, AB = L colk(A) rowk(B) == [-62 -3] + [ 4 · 5 ] [· 12 6]
10 8]
k=l 1 -16 -20 + 30 15 = [ 16 -4 ·
Oiflnltfon oclc up and lo-., trlang.,la~
,m=:::
a,:::.ilr=-
ce~,.--.~,.....,...,....,.~~lff'fflr,,l'WT"'>;-'"!'C~:lfflff."!'1
A matrix of a form A• [ ~ 1
tJ
is called block upper triangular matrix.
A matrix of a form B = [:: ~
· is called block lower trian matrix.
Inverse of Partitioned Matrices
I
The following example shows to produce an inverse of a partitioned matrix.

~ where Au is pxp, A
1
Example 3: Let A .= [ ~ 22 is qxq and A is invertible. Find a formula
for A-1• • l

Solution: _
Let A is invertible. So, A-1 exists.

Suppose A-1 = B = [ : : ::], then AA-1 =I= AB.

Here,
• • r

..... 0) .
This gives,
AuBu + A121½1 = Ip · · ·· · .. (ii)
AuBu + A12Bi2~=~0 · · .. · .. (m)
A221½t = 0 , ...... (iv)
A22°8.22 = lq . . .....(v)
. a square matrix.
Since A22 1S . So' bY invertible
· matrix theorem, · A22 is invertible. Therefore from
(v),
..... . (vi)
From (iv),
l½t = A22-l )( 0 = 0 ...... (vii) .l

Usmg (vii) to (ii) then,

A11B11 + 0 =Ip
- => A11B11 = Ip ' ·. . . 'bl
uare matrix Au IS ~verti e, so
By ~ e reason used to A22,A~~ sq , ...... (viii)
B11 = 11 '
Now (iii) can be written as,
, A11Bu + A121½2 = 0
-. A Bii = -A12An-1 [Using (vi).] ·
- 11 -1
• .I ' => Bt2= JA11-1 A12 An .
Hence,
Bt1 Bt12l
A-t = [ B21 8 nJ
 IIIATBEIIATIC&-11 ,

1
-[Au-1 -Au-1 A12 An- ]
7

- 0 . An-1 .

'
Example 4: . The Block upper triangular matrix A =
[An
o
~j where Att and. Arz are p><p and
dA are inventible.
q><q matrix, is invertible, if and only if both An an 22

Proof: Let A =[ ~
1
1:J is invertible
⇒ A-1 exist.
1 1 1
A•l = [Aoii- -A1i- Au A22· ] .
⇒ A77-t exist
⇒ Au-1 and A22-1 exist.
⇒ Au and A22. are invertible.
Conversely
Let Ap and A77 are invertible
⇒ Au-1 and A22-1 exist. ' .t

⇒ D-
-[An-o 1 -Au-1A12 A22.·1]
A22.-l
.
exISt.
1
⇒ AD= [Au 7
A12 [An-
1 1
-Au- A12 A22.· ] =[I 0] =I
0 An.J O A22-1 0 1 .,.

:_:I::v::~[yt~~li:f-J.~::~~n is ;tlo~~d~:~a~e.
0 0 : 2 '' -2 . ,_
1

A
:
0

. · '
Solution:
Set,

Then,
.,.. ..... '4" .. :' ••

Clearly,

A=[~u !j. ,. )

Then using the formula for inverse of partitioned matrix, we obtain

A-
A-1- ....11
1
:.A~; A12 A~] (')
''
-[ 0 A;1 ' ····:· 1
Here,
'. IJ
f

-Au-lA12A22-1 =:..[-2, l .3/2][9 OJ [1/2 1/2:J =[33/4

1 -1/2 0 1 . 1/2
-O -17/4
-2 3/2 33/4 9
Therefore(i)becomes,A-1= 1 -l/2 -17/4 -9/2
0 0 1/2 1/2 .
0 0 1/2 o
 llatrlz Alplna w l§i.irat31
Nunterical Importance of Partitioned Mahices
If the matrix is too large then the com utei - . . · · ·
-..-.or than the process to solve ma-.:~p .th partitioned and work. This proc;ess will give the solution
fCl-'~"" of the matrix.
inverse • U.lA WI
· out partitioned. Jbis
• c~ncept is also useful in obtaining an
I

' '
EXERCISE 3.2 1./

1. Let
3 2
2 3
1 5
-4-'-- i-
-1 2
-2--3
,The partition of A anci B is ~i\own. Then obtain AB, if possible.

2.
_..3 , 1.
Let A =[ 1 . -4 , 5 and B =
2J , ,,
I
[?~.' "j
e
rt"i ' t[ -
d : ~ind Ats by column ro~ expansion.
if . ., . ' ' .
• • • L

~nan:~ff '·i] :;~~~ ~

l • . • • ., j ~ a .

3
Let A ~ [ !s ~ ,- = applying column'row ~~o~ ~ ~od.

"I •· ,, 2 ~ • ,·. 1. '3 . r w:. i .

.. "1. i -2 · · 5 2 ·. :!
r · ·, _· FindiA-1 • 1 '
4 . If :Ai= ' I
0 0 .· ,4 3 ·.. .. . .
0 0 -1,1
2)• ·are inverses
. of each other.
5. Show that matrices A = ( ~ ; ) ~dB = ( _\
-
1

1 -2 -1] I

6. Find inverse of A =
[ -1
5
5
-4
6
5
if~xists.

7. Find inverse of A=[

• ' i )
i ~ ;]
4 -3' 8
... 1 • ' • I . I l
\ •. - j
. ...

«/

15 18] -3a+c+2e ' -3b+d+2f7 3

2. [ a_ 4c + Se b - 4d + 5£ J ·
[s· 1J l j
-7 1~
1.
••
[ -2 13
-44• I -36 .;
, .. 9 -1~ I 3 .
43 43 '43
i·.. · 4 32 -13
-3 13/7 -39/7] 6. A ~ not invertible.
7.A-1= ~ 43 43 43

4. [i -2
0
0
10/7
1/7
1/7
-29/7
-3/7
4/7 ·
. . 'I . ..
l '
-3 19
43 43
-1
43
J
 JIATJIBIIATics-11
"
3.S MATRIX FACIVRIZATIONS
, • A as a product of two .or more Il\atric
. tliat expresses " f es.
A factorization ot a matrix A is an equation . f d ta (combining th~ euects o two or more
. . Ives a synthesis o a . ti'o·n is· an, analys1s ~
. o. f uata.
Whereas matrix multiplication mvo In n.
. • · trix factonza . 1-1.,e
linear transformations into a single matrtx), ma od rh ~ounts"to a preprocessing of the
· th pression of A as a pr u"'-. · c..'~1 •
language of computer soence, e ex h ~ structures are more use1w m some
data in A, organizing that data into two or more parts w ~ _
way, perhaps more accessible for computation.

The LU Factorization .
1
m.nton industrial and business
by the fair1Y co
•
The LU factorization, described below, is motivated . · .
1
an
problem of solving a sequence of equations, . ~
• •th
' th-,
e same. coefficient matrtx:
<' ·
I

Ax= bi, Ax=~ ........, - Ax= bp ..... (i)

where the inverse power method ls used to es~t~ ei_gewalues, of a matrix by so,lxiug ~uations.

J When A is invertible, one could compute A-1 a.I),d. th~~ c;ifiput~ 1A-1bi,.~-11:12, ~d s<ion:.. ~owever, it
is more efficient to solve ' tne ·first equ'.atio~ in sequence. (~) by row reduction and obtain and LU
1

factorization of A at the same time. Thereafter, the remaiJ?ing-~~uations )1;t sequenc~Jl) are solved

' with the LU:factoJ¥clµqn. -= •• : ·1 c-.

I
'.1.J', .·., . ~i . _ · , ,
I
..
. '. . ~·- ! : 1 • :
l -

At first, assume that A is an mxn.matrix that can be row reduced to echelonn

' , form,
.
-without row -
interchanges. Then A can be written in the form A= LU, where Lis an q,.~m-lowerj:riangular matrix
with 1's on the diagonal and U is an mxn echelon form of A. Such a· factorization is ciilled an LU
factorization of A. The matrix L is invertible and is called a unit-lower tri~gular maW. '
' . .

A= *·[! '!0 ~0 OJ [■g· ~* ·~-

f *·\~:·* :,:*]<· ·;_ '
* • • 1 o o o o o· f
r r ,.
L u
Fig. Ah LU factorization. ·. (.
' .
Before studying how to construct L and U, we should look at why they ~ :so useful. When A ~ LU the 1

equation Ax= b can be writtep as ~x)_= .~· ,Wtiffl:lg f ~r V~ w~ ~ find x pY ~lving the pair of equa;ons.
Ly:=b

'. Ux = y
First soh,:e Ly= b_f~r y, and then solv~ Ux = y for x. iach equation is easy to solve ~-L .-. - · U
. guiar _ _ IA:\,;ause I: and
aretrian . - ,
Exaqtple 1: It can be verified that
I
. r ..
- •-
,- [3
- -3
·-7
5
-2
1
21· [1-1 ,.~ 1o ·o·0 oO
0
3
_,
-2
0 -2 " -1
2
2
'1

A= '6 -4 0 -s=2 -s 1 0
-9 5 -5 12 -3 8 3 1
0 0 -1 1 =LU
0 0 0 -1
 . , Multiplication
byU
.
·-------
Y Multiplication
. '

p· by L
ig: Factorization of the mappm·g Ax .
X➔

' Use this LU fadorizati

. on of A to solve Ax = b h [-9]
5 .
• , , ,wereb=
1
. . 7
Solution: : .. ~ . ' 11 ; ·,
) & • I •

Toe solution of Ly = \> needs onl 6 : . .. . .

· , , Y multipli ti
place only •in column 5. (The zero ~low eachcapivot L 6 additions, because• the an
91'\S •and · 'thmeru;
. takes
·
5
row operations.) m are created automatically by the choice of

-
0 0 0Q;~. Q . -9]'- r.

1] -[~
Q Q ·- fI
0- 0 If -4
1 o. 0 1 0 . 5 = [I y]
3 1 11 O 'o o( 1 -1
tiout row
Then,.for . =. y, the "backward" p h ase of -row reduction
. Ux · r
··. , - 4 di . . . . . -
tr matrix
6 additions, (For instani:e, creating the zeros'in col ,.,4 twres "'."'ons, 6 nrul~plications, and
danLU multiplication_.,additit>n pairs to add multiples 'of· umn4 oth[U y] r_ equrres 1 division in row 4 and 3
. row to e rows.above~) • . ..

=[~ =~ =: ; .1] -[~

0 0 0

[U y]
0 0 -1 1 -6
1 0 0
0 0 1 0
!l,x=[!l
_{j _{j
0 0 0 -1 -1 0 0 0 1
To~~ x requires 28 arithmetic operations, or "flops" (floating point operations), excluding the cost
of finding Land U. In contrast, row reduction of [A b] to [I x] takes 62 operations.
'
Algorithm for an LU Factorization
1. Reduce A to an echelon form U by a sequence of row replaceinent operations, if possible.

2· Place entries in L sµch that the same sequence of row operations r~duces L to I.
. ' .
Step 1 is not always ·possible, but when it is, the argument above shows that an LU factorization
exists. f,x-2 W;ill show hqw to impJ~t step 2 By constructi~n. L will satisfy . .

(Ep ....... E1)L = I

USing the ~ E
3 .
~
_ . Eq ( ) Thus L will be invertible, by the Invertible Matrix Theorem.
With m E) e 1 •.••• ~asm
U d A = LU So step 2 will produce an acceptable L -
\61> •••• ~ -= L-1• From (3),,L- A= , an
1
·
 ilATBBIIATICS-0 .

Example 2: Find an LU factorization of

!~
0

A=[1 ; ;]
8
1
-6 0 7 -3
Solution: f A d. .d d
Since A has four rows, L should be 4><4. The first colu~ of L is the first column ivi e by the °·
top pivot entry:

L=[-; ~ g g]
1 1 0
-3 1 · ·• . ,,.
Compare the first columns of A and L. The row operations that create ~eros in the firS t column ~£ A
will also create zeros in the first column of L. To make this same correspondence of row opera!Jons
on A hold for th; rest of L, watch a row reduction of A to an echelon form U. That is, highlight the
entries in each matrix that are ·used.-to detenriirte the sequence of row· operations that transform A
into u. 1 I ''l f • • I •

. .,
4 2 -1 5 -2] [2 4 -1 5 -2]
-5 -4 3 -8 1 0 3 1 2 -3
A = 2 -5 -4 1 8 ~ 0 -9 -3 . -4 10 =.At ' ~
[ v

r~ (-{ ~(:~]
-6 0 7 -3 1 O, 1 12 . 4 ., 12; -,5 ~ ;:;

i;. .::; J •
~Ai=[~ ; ;1 ~ :;]0 I O O 2 ,l < '": 0 I Qr •/_l~ • 2 •·• 1 ' ;==
:=-
u .
I

_,·.. :
"!',-: ·~:.2•-· !/:
_; :
· ..:r._.•• :,~.7,' ' - o. 0 0 [ ·i .z ..
ll ' O'.... ~, 0.,,-, 0 J'.) 5
,IQ •, _;f •
1 ..
...J
•
l..1 .. · .
1
•~r•\,•

, i·.~-=..; , ...,:: The highlighted entries above determine· the:row·redu~tion,of·A lto U. At.each pivot column; divide
the highlighted entries by th~ pivot and place and result 4',to L: ·

1 l [-!] m
[-6j 12
rs1 : .: :
l·
: . :_; ;: ,
. ,,
. :. ~ ::
,_ _
0 : •
.,

. +2 +3 v,, +2 +5 I '.. I 0

!1 ! , ,! , !
_? f zi :, } ..

[
-2
1. i
-3 4
1
-3
-2·..
'
J. mdL•[! -r-.i --il
An easy calculation verifies that this L and U satisfy LU = A.
'
J' . :

.... . .,
Numerical Notes: f I. • • I ' ,. • •• ··~!
0
1 II' t I !l "1 f'j 1 j
The following operation ·counts apply 'to an nxn de~ matrix . , . , ,
moderately large, say, n ~ 30. . A (with ·most entries 'nonzero) for n
1. Computing an LU factorization of A takes about 2n3 1 • •
· ~ [A b)), whereas'finding Ar.t requires about-2h3 ff
3
/ flops (about the same as row reduciJ\8
2 Sol . L b d ops. .
· vmg Y = an Ux ·J:? Y requires about 2n2 fl ·' ' ··
solved in about n2 flops. ops, because any nxn triangular system call lt
 .,
fdultiplication of b by. A-1 also l'eqUin!s , llatrt.~ W§r i'a31
3· 1,tained from L and U (beca about~ flops, but the result
0 USe of roundoff may l\dtlbe u llccurate as that
A-lb~- . .· ~ . • . error i when ' computing both A-1 and
4. If A JS ~ (with m~y z.ero entries), then ~ U LI • , •. , rf .
dense. In this case, a solution of Ax -- b wt"th an LU factoriza
may. be .-r-~,
!lrn:I..... too -w L - - A 1 . likel to be
' 'a::rea5 - JS Y
' a tion JS,pwchfaster than . A-1
THE LEONTIEF INP · ·'· •.,;. , . usmg ·
3.G tJT OtJTptJT MODEL
~

And, another part of economy of a nation is -d b - ·· · · · · -

· th . _ covere Y consumers. Some consumers used the
goods to. consume emselves and sometimes some prod U\.uon~ · -0 · •
nouses· use me.
~o · -- ,
goods lor'Uteir own
production. . ,; r.,. . ...
~ t ' • • '

De11nlllon (
The total
denwuj .v:

Definiilo
tbe,addi
L, dividf
theirown . .
' . .;. f.

On the real world, the_interrelations betweens varies sectors are var} complex. Therefore, the
production and demand is difficult to study. HQwever, Leontief asked that the production level will
exactly balance with total demand (sum of in~ermediate and final al m~d):'~t'is, I\ \ _1
• • - ' t • • , l • •• • • 1 ' ) • ,. t l ; ~ ... •

x = intermediate demand + d. · ' : ·

,.

,- .~ , -,:..,

Example 3: A survey-report s h ows a data

I fot~ Inputs consumed per unit of outJ>ut
Manufacturing Aericulture . Service -
Purchased from ..
0.4 0.2 : ,

Manufacturing 0.5
0.3 I 0.1

Agriculture 0.2. ' V •

I
0.1 0.3
Service
0~1 .
 r

IIATIDUIATICS-U uf turing ·sector if it decides ta

b the roan ac .
(a) What amount will be conswned y .
produce 100 anits? - ~ if it decides to produce 20
d by the servfce ~ ......
(b) What amount will be co~swne ..
'. units?
,. r
f • J • '
.
(c) Find the intermediate demand.
Solution: • -~ _ c3 be _unit-output by and
Let c, be unit output by manufacturing, c2 be unit-output by_:agn · ture_ - .
• t... J •' .. •
service. . . I ')

(a) Given that 100 unit to be produced by ~ufact_urinS· : 1 • • .1...J • : • • : •

1
u

[:J . ~·
1

So, lOOc1 • 100 [~:~] =

0.1 10 . ·ts from
• .· . d d nsume 50 uruts, 20 um
This means to produce 100 units, manufacturing will or er an co . .
agriculture and 10 units from services. i· " .•

(b) Similar to (a), where use 20.-~tead of 100 ~d c3 ~~ad of c1.

(c) Since the intermediate demand is x1c1 + X2:C2 + x:3<:3. So,

Intermediate demand = c1x1 + c2X2 + C3X3
=Cx.,' J • t.
where, . .'
o.s o.4
l
0.2]
C = 0.2 0.3 0.1 = (c1, C2, C3)
0.1 0.1 0.3 ·
• J • • I

and
.,
'I.
1. ! J
Definition (Co
The matrix nota
Example 4: Consider the production model x = Cx + d for an economy with two sectors where

C a[~:~ ~:;], d • [:J . ~

' . . ..
Use an invene matrix to determine the production level n e e ~ .to satisfy -the final
deinand. .
Solution:
The model is,
x=Cx+d ' .
⇒ · (I -.qx ., d
• ► .~ •• (i)
Since, • f

C=[0.0 0.5] and

0.6 0.2
 fhen (i) can be rewrite as, lletrtaAJaebra w §ria:J

[~~ ~:TI[;~= [;gJ

'fhe augmented matrix of (ii) is,
1
[ -0.6 -0.s
0.8
so]
30

~[-0.61 -1.2
8
SOJ
300
r

[Applying R2 ➔ 10~]

-[~ -1.2 :J [Applying R2 ➔ ~ + 6Rt] ,

~[ol -1.2 SO]
1 120 [Applfhli R2 ~ R2/S]
~[ol 1O 120
110] [Applying R1 ➔ R1 + R2/2]
I

Thus, th~ production level is,

. =[xil
XxJ =[110]
120 .
Example 5: Consider Leontief input-output model equation x = Cx + d, where consumption
ma~.C ~s . . i ••..•.• ~· .. . 1•
I •I ~ •
·. . C =·[.so
.20
.40
.30
.20]
.10 . .
.10 .10 .30
Suppose final demand is 50 uni~-manufacturing, 39 unit for agriculture and 20 unit
' , • .,, t f ~

for service. Find production level x that will satisfy this demand.
Solution:
.40.so
.20]
Giy_en ' = .20 .30 .10 , ~-
[ .10 . ..10 .30
- . ' . , ,11 '

Since, x = Cx + d .
--
rr ,rn 1, l,,ut

-.,
hert Hence, (I - qx-= d ..... (A) I
·~ 'J
Here, -- -

-
e) R ·1(:so
1-C= 0 .1, .., 0 - .20
.40
.30
.10
,.10 = cso:
·.30
- .20
- .10
-.40
.70
-.10
- .10,
.70
-0 0 1 .10
Solving (A); Augmented matrix is .., '

-2~ C ~
-.40 -4 -2
l•
c50
- ..20 .70 - .10 ~ -2
.70 -1
7
-1
-1
7 200
•. 1
. ~

-.10 -.10

+:2
1
7
-4
-7
-1
-2
-27 300
500
I j I

r
-' r
j

1 -7
~ 0 9 -15 -200J
-100 I •

0 -9 33 1500
 MATHEMATICS-II
~
1 -7 -200)
-100 ·,'
-rn
9
0
-15
18 1400 .
I

.J

1
1
-7
-5/3
-200)
-100/9
.. t, ,rJ .J,; ,. '

-rn 0 18 1400 ,. •c f • ~ f).j

0 -16/3 -1700/9) ' J

-G -5/3 . -100/9 r
1
0 1 . ,700/9 . '1 • .

0 0 6100/1 r•
' _t.,
r .!
-G 1 0 92ffJ/,+7, . :
'

,
J

0 1 700/9
0 0
1 . ·o 119'
• f • • .' I f

-rn
~

0 1 78

lff
\I

' , r, , . J r
•r
·-
Thus production level X = 119 Manufa&re
Agriculture
78 Service

ition (Column Sum

I
~, '

r. . l ·column sµm be

Theorem 11: Let C be ·

H C and d have non:..
exists and the Prt?ducti
' .-
x =
~

,.
t
·non-negative en
.~~~A .if• -;~.
· , x ,,= .C x+d; -

A Formula for (I - C)-1 r·, 1

, : r--

Setting the production level is same as the final demand: a¥ipitial year. So, x ·~ d initial year. for
Therefore, 6 = Cd be the intermediat~ demand of fir~t round (first year) then input needed to meet
the demand Cd is C(Cd) = C2d. The demand of continue process is as, ·' · · 1•
Demand · In uf ~
1 round
st

2nd round f •

rd
3 round
and soon.
Since the production level x will meet the demand. ~,
X = (d + Cd + C2d + ... .. . ..) I
- -I
=(I+ c + c2 + ...... ) a'' \
....-.. (i) f. I •
Since,
 .
If the column sums m care strictly less than 1 then
........
y
•
I -r-• ••·
m

ri,eretore, (ii) implies . 1 - ,C Js invertible. And, C- ➔ 0 u m➔oo.

(I -- q (I + C + C2 + ... + €fn-t + Cm) ~ I

asm ➔-
:::> • (I - q-1 • (I + C + C2 + .. . + ~1 + Cm).
provided that th' column sums of c-are 1~ than l.
F,conomic Importance of Entries In {t - q -1
The entries in jtlt column of (t - ,.,('_, th J_ _!_ _
. . '-J are e uicrease~ amounts the various sector will have to
produce m order to satisfy an increase 0 f 1 . . · . ' · •
• • urut m the final demand for ·output from sector j.
Thus, the entries m (I - q-t can be used to predict how the production level x will have to
change when the final demand d changes. ,
Numerical importance of (I - q-1 ' I

.. .... . .._' ,. l r.
j, •
, • \ - 4 (Jf t ~ I • t

H the 1sy~ ~ 1= blµ; ~ge and ~ ~some non-zero entries, then (I- q-1 will provide practi¢
formulas for solving Ax = b with A = I - C whenever the columns sums of the absolute values
in ~, ~}~ than 1. ~ 1helps}~fin4ing A-t. . . ..
3.7 APPUUTIONS OF MATH~ ALGEBRA mp t;oMP1JTEn GRAPa1cs

Computer graphics are images displaye&br animated a comp11ter screen. Th~ mathematics on
used to manipulate and display graphical images. Such images consists a number of points,
connecting curves and inform about how to fill in the closed and·bounded r~gion by curves.
___..,. In the following example, the effect in -picture is shown by using shear transformation.

r initial
!eded to
Example 6:

Forgiven
The coordinat~ of vertices of the .adjoining figure are (0, ·0), (O.S, 0), (O.S, 6.42), (6, 0),
(6, 8), (5.5, 8), (5.5, 1.58), (0, 8) in order.
•

NS .
8

3 \
I

As[~ :5] 0
1 2 4

describe the effect of the shear transformation x➔Ax.

Solution: Consider the coordinate of figure-1 as in ~bix form,

~
55 8

o = [~
0
,j5 :,~ : 8 :.is ~]
Then
,:, , r,o o.s . z.105 6 ._8 7·5 5.895 28] . ·
58
.. ; AD - ~O O ~~ O .8 8 1.
. . th fi e will take new form.
This is the effect of D by A. Therefore, e gur
 MATDIIATICB-11 1 v J rn
. · 2D and 3D l r.lill1 . . ,
Homogeneous Coordinates m .-.cmond directly to matrix ntultipli .
does not cou-....r- "d this ..a:u. <:atil>t\
.
Smce the translating an object on a screen . The process to avoJ . '-UlllcuJ.ty is
. tranSformatton. ~
because the translation is not a linear -..
• . C ' - '
homogeneous coordinates. · l) which is homogeneous COOr<ihla_
. tifi d · R.3 as (x, Y, ~
.-,,1:..... te5 (i , y, iz, 1)? The homogenen..
2
Any point (x, y) in R can be 1den e m
. h h geneous coo1u1.ua , --~
Likewise, any point (x, y, z) m R 3 as omo . nix ,.
. will be seen m new
coordinate is not use for matrix algebra but It , 1
ma . ·
•
•
•

• • in homoge~eous coordinate,s as (x, y, l)

Example 7: The translate (x, y) ➔ (x + 3, y + ~) 1s wntte0
• ➔ (x+3,y+4,1).Inmatrixform n ., , _,, r. .' -::. ~; •-;u;

- .c . r. rn l :J[rJ=[:;:J ... 1· .••••• -·.:.· • G'

Composite Transformations ; ·~-i - , • , ,

or
J For the movement of a picture on a computer screen often re~uir~ two

when homogeneous coordinates-are used.

•
. .
·1 . r I
more
transformations. The c~~position of such transformations corresponds to a -matrix• multiplication
J

I Example 8:

~- · .

(i)
Any linear transformation on R2

coordinates by:a parti~o~e~ .m~~-~f

Some examples are
.Counter cIockwise rotatiRn ~~ut the origin angle 8
is -repres~nted

.. , ,
witli ·;~sped to llom.ogeneo111
~/j~~-.[!..:. ~], wh~~ A-~ a 2x':_matrix.

0
~[;~~ -~::e ~]
0 1
I k ·,, ,
.: i ' J
(ii) Refl~on thr~ugh y = x.

.~ · r~·i ., .
..,
Lo o.
(ill) Scale x ,by s and y by t
0
[g
(iv) Translation (x, y) by (x + h, y + k)
t
0 ~] '
.. l

'
t• '. ~

. /1/
, /
/'~ I
I
'[ :~
0
f
0
t]
1
Example 9: . Find a 3><3 matrix that corresponds to th . .
0.3, rotation of 900 and finally a transl ti e composite transformation of a scaling by
. a on that adds (~ S
Solution: . • , 2) to each point of unage.
Here, (x, y) ➔ (0.3x, 0.3y) be the transl .
homogeneous coordinates IS. ation. Then the matnx
' form of the translation
 llatrlaAJpbna m IC&Ama3'
[n3 -0

m ➔ _Q
0
0.3
0 ~][t] (scaling)

➔ -rn -1g 1
~] [0.g3 o0a 0o)[;]
0 (rotation)
0.-0.2-S~tO1 -1 g][Oo3
11
-➔ -rn o_ . t o o t o ❖
-
1 ~ ~)[i]
}', l)

-/,, ·J- rJ[rr5].r[•3·m

11ie matrix for the composite transfonnati
f . . (translation)

r ~ gJ
r onm •• ,

i-:_; ."'-.'..'. _. ·
1 ,1 •, j O 1 Jo, I
~
' [() I 1• ' • •••

·= 1
0
o0 -2 ·0 0] 1 [ '~ , , .3
0=.3 >- 0
,:-2.s]· '. ,. 1 ••

nore t o o t o· o t , ,
tion
3.8 SIJBSPA£ES.8F fil,N
, ([•• 1 •

This section focuses on important sets of vectol.1jl-

:.._, m
: Rn called
· ' -subspaces
-' ·,;v---
· · ·'
subspa · ·
trix. ~~~~~ ~ -~'
wm- n WI ~matrix A 'al\d they,provid~·usefulrinfonhation about the equation Ax= b ..The
~~m
concepts and terminology in this section will~l,e·used repeatedly throughout the rest of the book: I

set H in Rn that ~ ~ ee Pf<?~ rti~:

(I r' • r1 ,,-L . • t , , I, , . , 1
Asubspace of Rn is any • • • I
. _ . '.
·' ' '
.• I •
. • •
r I
•11

tflte -zero vector is'in H. . • . ,.~~ ,. 6.

1
'i1' r <
1 I .,. ., •
t •·

•.. ◄ ~
..._, .. ..J
. . . 1.. • • •• . . .:•;. • • • J .. '· '. • :..' )

b. i Fof ~~ u an~,v;-~
~, ~e.. ~~~u h~:±.v is in H. ~•·. i'
t \ ' { .
c. 'For eacfi ih Hand· each scalar c; the \iector cu is in H.
. Ii

In words, a su~pace is closed under addition ~d scalar

' I, •
multiplication. As you wiµ see in the next
• • I I l I • • .,,.

few examples, most sets of vectors discussed are subspaces. For instance, a plane through the origin
. I' ( I , , . , 1ft , l ' • • ; • I
IS the standard way to visualize the su~pace in figure. ·
Example 10: If Vt and V2 are in R• and H =: spani{Vt, V2}, then His a subspace of Rn. To verify this
· statement, note tha{ the zero vector,is pt H (becaus, Ov1 + Ov2 ia a_linear__ c~mbination
of Vt and v 2). Now ~e tw~ at~itrt"Y vectors in H, say. ·

u = StVt + 52v2 and v = tivt + bv2 l •

Then, I .
11
U +V = (St + tt)Vt + (52 + b)V2 • - ~:
which h . . b" n·on of Vt ana v 2 and hence' is in H. Also, for any scalar c,
s ows that u + v IS a linear com ma , ,
2 2
th e Vector cu is.in H, because cu c(stVt + s2v2) (csi)Vt + (cs )v · = . =
If d, iJnp\y span a line through the origin. So a
vi is not ~ and if v is a multiple of vi, then vi an v s
2
·· ·
line through the origin ~ another example of a subspace.
2
' '
 ~JIATICS-0 e b~cause it does not conttan \l.
11A·1p.u ot a subspaa , . 'd dditi • "ll
th origin is n . ot closed un er a on or 8(~
. A line L not thrOUgh e F'g 2 shows that L 18 n ~ ,
ExanlPIe 11• o o as rolllU&,..
--~-d• A}so, l • , .
')
.
I t
ongtn, a ... , •• 'l I

multiplication. ~

'. I

I ... I L
I.. '. 2wiano(OOL
· • + ••-aaL . f,., . · Vp is, a ,fJU.bspac~ of Rn.
, , . all lin ar coJ]'lbina.tiOnB o , . i . i, ....., ·,
Example 12: For v1, ....., vp in Rn, ~e set of . e imilar to the argument given in £~ple 1. We
The verification of this statement 1S s . ' d (or g'enerated) by Vt .... Vp-
) s the subspa~e spanne . F
shall now refer to Span {v1, ...., Vp a ~11 _1 rties .required ror a subspace. 1

Note that Rn is a subspace of itseIf beeause 1

·t has the uuee prope
, tor in Rn. This set,
called the zero
Another special subspace is the set consisting of o~~ the zero . ~ . .
subspace, also satisfies the conditiorts for a subsP_ac;e. , !i r 1
I lI

3.9 DIMENSION AND RANK

~1ta; ~~.:~
. ~ry
, It ~ be shown ffiit if~ su~ pace ;-1,~i~oiJ, ; ~ t5'~
exact:Iy p vectqrs. 'f.hus·the follQwing-definition make5isen,se,- "
-b~~-ofH- ~ ust consist of
' ·.,: v: ·• ,... , -~- : · L ~
The dimension of a nonzero subspace H, denoted by dim H, is,the nwrtber of. vectors ihiany, basis for
1

H. Thepimeqsipnpf-thezerqsu~P.ff:e{O}~_defjn~gJo,be:~ro . .1 ,-• .,:.-, . ·

1

:· ., .,., . . . 1,--' l'-J 1•

The space Rn has dimension n. Every·basis for Rn consi~ts of n vec;tor~. A. plane thro_u gh O in R3 is
J •

two-dimensional, and a line through Ois on~aiinensio:rlaf. ; .t_·· 1• . ' . · . • ·, ·•. • , ;' 1 .-~ ',

Example 13: Recall that the null space of the matrix A had a basis of 3 ve~o~. ~ th~ 4in;)gnsioq";
of Nul A in this case is 3. Observe how each basis vector corresponds to a free
f ,• __ . f J ( I -

variable in the equation Ax = 0. Out toristrucliod··always 'produce~ 1,aiis 1ffi this .

way. So, to find the dimensio11r .o~ ~µl ~ simply i4,en~ an~ ~0UJlt t\t~ nqmbf!r of
free variables in Ax = 0.
Th~'i-.r-a.J{ of a ~trix A, ,denoted by r~ llted~ ~~:oi{ dftl\~,~~i~ space...ciff.:• . ... ~ - '
1
A, is
• ,1
1 , I ,,,., ~ ' 'J'1~ . J - ' , ., • ~~,, .l,"-'l 4,,.,. •,, •<".; • ,. ,..~t.- l_,,.1 '
Since the pivot columns of A form a basis for Co~ A, the rank 9f_ A is jus{ 'the· ii~~ ,.~~ pJ·lVO
. t
coIumns m. A. -,- .,,. . ft • . ·• .,, ' . ...;:.t ' J. •
•. ' •
, . . ::... • • 1;,•.: 1 ~ "-:. ·.'.

Solution:
Reduce A to eachelon form: - ,.

2
0 -35 -32
-4
5 -7 0 -3 2 5
-4BJ [2 5 -3
A~
1
•

The Rank Theorem

i
0 -64 14 -20 ~· .. · ~ 0 0 0 4
9. ,-~ •_6 ,5 j ,-6 . ( l l O O , 0 ,, 0 .
The mabix A has 3 pivot columns, ~rank.A,._ =
I • •
3.
·•--! . ., · 1 • .i .1 • · ~ , •• •

H a.;..,matrix knas 'n columns, then rN:tik A·+ dim Nu.I A"= n
1
-, • - , ,
whe~e .~- Nul (A) ·= number of non
pivot columns. . ,., . r I ' I
, . ' . •• ! J.,. I
r
 fhe Basis Th~rem .
llatrla AJae1,ra ID ICaAmR 31
1,et,Ef.J,e•Jlp-:dim~nal subspaq!<0f Rn. ·Ah •linear . . . ~ .. . . .
tolJ18tically a basis for H Also, any set of p e1 y ly mdependent set of exactly ·p elements in H is
au •entents of H that -nco H · • · basis c
fhe 1nverti"ble M ahix Theoreni -r-~"" 1S·automatically,a , 1or H.

Let A be an n><n matrix. Then the followin ta ·

an invertibJ~ matrix._u~b .. , _ l: • ~ ,g s tements are each equivalent to the statemen~ that A is

The columns of A form a bas~.o~Rn. .,

• • •~ . 1•
ColA=Rn r

dimColA=n
'.
rankA=n
NulA = {O} I
I
'

dimNtilA=O
Numerical ~otes 1 •• ., . ... .
al I ~:a.L--- ~:~-·---..JI . this .
Many ~~uwi:;.. U l . : K : ~ m text ~are useJ.uI for.,· und~ding concepts and- ~ g s~ple
computatlo~-by ~~- However, the algon~ are often unswtt1ble for lar~e problems m i;eaJ J.ife.
Rank deterinination IS a g~d example. It.would seem.easy to reduce a matrix to echelon form and
count thl pivots. But unless exact arithmetic is performed ·on a matrix whose entries ~ .e spec;ified
exactly, row operations can change the apparent rank of a matrix. For in$tan&, ifthevalue bf X ih the
matrix G~ is not stored exactly as 7 in a computer, then the rank ,;,.,~ be'~ or. 2, d~ding on
whether the computer treats x - ·7 as zero. · . \.
In practical applications, the effective rank of a matrix·A is often determin~_drfr~~ ?1e ~in~ ~alue

• ; - 1, 1-... ~ . . , ·~ ..

1, Co "der the production·model x =:= ~ + d for an economy wi~ tw~ sectors w!1er~, __,,
rb.1 '0.6] andd--[18]
C=Lo.s o.2 11
. : . ~ -.-. . . . ; . ·.
,
· . trix to determine the production level necessary to satisfy the final de~d. t
Use an inverse ma 1ili thr tors uiven that · · 1

2 Sol th Leontief production equation for an economy w1 ee sec ' o-

C=v[i~eg:i g:~]andd=[:] . .. '- ; .

80
0.1 0.0 0.2 · late (x, y) coordinate by (~ ,) and then rotate ·45° about
3. Find the 3x3 matrix produce the trans J I I ,, . : .
the origin. . = Cx + d for an economy with two,sectors, where
4. Consider the production model x .. ·'

[
o · 0.5] -[501 - : . ·
C = 0.6 0.2
d - 30JI . th duction level necesSM}' to satisfy the final demand.
Use an inverse matrix to determine . e fr; an economy with three sectors, given that :_

5. ::v[egr !:f~f]od::ti:: eq[i]ation o · _

0.1 0 0.2 so
 MATBBMATICS-U

Given the Leontief input-output model . . . '

6. x = ex + d, where the symbols have their usual mearuns- Constder· any econom.y
consumption matrix is given by "'ho..
0.5 0.4 0.2]
C = 0.2 0.3 0.1
[ 0.1
Su"f""" ~ 0.1 0.3
final demand is 50 units for manufacturing 30 units for agriculture, 20 units 1
services. Find the production level x that will satisfy this demand. llt

11 3]
7. Find the rank of a matrix A = 5 [2
2
2
4
6

8.
.
Find the dimension of null space of matrix A = 2
[1 22 ,41]
3 5

9. Find an LU factorization of the following matric!s 3

2 -4 -2
-5

(a) [i -1
-2 2]
10 (b)
[1i 3
-5
·2
' 8
-5 11s·
-7
I
(c)
I 6
2
4 .-2
-9
-7
-5
-3
-;-2 . 7'1
8
9
-5 6 7 .. -

'
-2 -4 -6 3 3' 4

fa~:on~~]
I

.. .....
10.
rm; L~
-4 · -5 ., 3 -8·_ 1 ·, I ·1
·.
, .:,~
[ I

I lF
•

'
2 -5-41 8 .{,·. . ... '
\.
·- - .. l '. l!
-6 0 7 -3 1
1 ,· n .... · 1 ..:, ~, .. .. " .. , ._1i; I tit
11. Find LU fact~riz.ation of [ 2 ' 5 ]
. 6 -7
2
12. Find LU factoriz.ati~n of [ -3 _ss ]

; !J.'
0
·-~
'V"-
11

4.• , • 120 .
[1101
. 8. •. O• .

:.J .

. .
1
12. LU • ( 3 ·O]
1 .I, I u. - [ 20 1
5 ]
-22
I<.
 DETERMINANTS
Oeterrn "~' - ole ,in the study of eigen
value::
, ; ;. . x, "~
::;'field of science and
engine ' :' ~-rtn~tion cha.nge~the
,.,, ik -,,., ;;,. ',f: "J•

area of ctfig,

LEARNING OUTC0~ES

After studying tbls chapter you should be a~le to:

~ Introduction to Determinants · II

& Properties of Determinants •

• ar Transformations
'&. Cramer's Rule, Volume an d Lme
l •
 11ATJIBJIAT1CS-II

0011cnoN TO DETERMINANTS
4.1 INTB _....,nes the value of the matrix. lh\ls
tiix A dete,uuu 2 2 nix ,~
d terminant of a square ma . li ity we consider a x ma ,
Normally, the e t of a matiix- For sunp c ,
determiJlant is a scalar coIDponen
=[au au1
A an an.J
Then its detenni.nant is defined as
au a121 = a11a22 - a12a21
.• det (A) = IA I = a 21 I a22
Likewise, the 3x3 IDatrix is
- :A= [anan aua22
cl31 cl32
and its determinant is
au a12 a13

.
det (A) = an a22 an
cl3t cl32 cl33
= a 11a 22c133 - a 11a23c132 + a 12a23clJ1 - a12a21cl33 + a13a21cl32- a13'12:~1 ~,..,.--.--.....----,,.----
• • • . ,. - ~"'V, ~. ,- ~ --~: ,::-¥.'"·71:: ·r,· ~t~fr ~- )/'# ~ - · ,
Definrtion (Determinant) . . · '>' ·• ;,,,. ..., , ,
For n > 2; the determina~tof ~ -nxn.Jn ... ') . .

I
I

- det(A) ~·n -; ·};, /,):s1~:Jj .

I

and range is a
r
Alternatively, determinant is a function which domain is a set of square matrices
subset of complex number. '· ' .,: :

Example 1: Comp~te the d ~ t of_,\ =H ~l -.. ... :' :

Solution:
Here,

det(A) = 1~
0 -2 0
! -~
= au det (Au) - a12 det (A12) + a13 det (A13)

= >
1
c I-~ ~I -csr I~ 11 I~ ! I
1

= (1) (0 - 2) - 5(0 - 0) + 0
+ <0>

=-2

' .
~I =-2.
.., • .J • • '"'i ' •.
To find the value of a matrix, next one IIJ.ethod is also useful .J •
Let A= [ai;J be a matrix, the (i, j)th- cofactor of A is de t db . .
 Determbrmta w fci#ma◄
such conce t leads the following theorem .
• TfuaQ.(11111 JI! The ~ t of an nx •
across the fdi row as · n matrix A c~ be computed by a cofactor expansion
det(A~ = a11 Cn + a12 ,...__ + + a ,..._
where, '-12 ••• 1n '-in

0 n . Cti = (- l )i+i det(AIJ),

The cofactor expansion across . .
th~ .11i coumnIS
, ~. , 1 1
. . det(A) = a11 C11 + a21 C2i + ... + ~ C
where IS defined above. · nj . nl
The following example helps to und tand th ·
. t. ers e cofactor expansion is also helpful to determine the
determman . , . ,
Example 2: Using cofacto[r exp;sion,] c~~pu~ the dete~ t of A wh~re .
1 0
A= 2 0 0 .
3 -2 3
Solution: 'L- I

Here,
det(A) = au Cu + a12 C12 + a13 C13
· = au(-1)1•1 det(Au) + a12(-l)~•; det(A12) + a13(-1)1+3 det(A13)

o
=1 l-2 ol
3 -5 3 12 ol3 + 0 12;3 I
01 ·
-2 •
=0-30+0
=-30.
Theorem-1 is helpful for computing the dete~ t 'of a ~ atnx tJ;lat c~mt~ many zero rows. This
:a means if a row has mostly zero then the cofactor expansion across that row. So that the cofactor in
those terms need not be calc~ted an~ same .process-works with a column.
Example 3: Using cofactor expansion, compute- the dete~ant of A where

A= 2
1

[1 5 0] 0 0 .
,
3 -2 3
Solution:
Here, second row has most zero. So,
det(A) = an Cn + a~~22 + a~C23 , . . :.
= a 2i(-1)2+1 det(An) + a22(.:.1)2+2 det(A22) + a23(-l) 2+3 det(A23)

=-21~2 ~l-o+O[·: a22~0,a23=0]

= -2 (15-0)
.= -30.
5 -7 2 2
0 3 0 -4 .
•t
Example 4: Compute -5 -8 0 3 , :- ..--~ ": 'l
t

0 5 0 -6
Solution:
Here ,--: . ,r1 . 'U ,..._,· .~ It I I
• I

5 -7 2 2
0 3 0 -4
-5 -8 0 3
0 5 0 -6
 0 1
0 3 -43 (".' third colUJI\Il has most zero, so
... (-1)1+3 (2) -5 -8
0 5 -6 e,cpanding from third column)

....... n has most zero, so

= (2) (- 1)2+1 (-5) 135
-41 . . 5eeond
-6 ( .
.
col
from second column)
U,U,OA .

expand mg

= (10) (-18 + 20) '> l

=20. u
. d • ant of A where
Eumple 5: Usmg cof:[T!loro1puT]·.. eternun ' .'

0 0 2 4 -1 1.
0 0 0 -2 0
Solution:
Here, second row has most zero. So,
3 -7 8 9 6
0 2 -5 7 3
5· · O

I
det (A) = 0 0 l
0 0 2 4 -1
0 0 0 -2 0
2 -5 7 3
0 1 5. 0
= (..:- t) 1+1 "(3)
I 0 2 4 -1 •. k I

.. . Q 0 -2.-c-• P. • :. • •.., .. I I

-. L :!?., 9
')· t
= (-1)1+1 (3) (2)
1·
2 4 -1 ..
0 -2 0
."
= (6) (-1)3+2 I~ ~11
= (-6) (-1- 0)
=6

Example 6: Find det(A) where A = [i ~

1 0
Solution:
Given matrix is a triangular matrix. So,
2 0 O
det{A) = 3 5 0 = 2xsx1 = 10.
1 0 1
 c-Jore the effect of an elementary row ·tt
l>etenmnanta ID §§41
~ l e 7.:. a.-r fa b 1 a r b1 opera on on the determinant ~ a matrix
Le dJ' Lice kdj · .

,~ ~, =ad-be
.,
and . I
a
kc kd I
b' = kad - kbc = k(ad - be) = k c la bld .
The determinant is multiplied by a scalar k is same as one row of the determinant is multiplied
by the scalar k. . .

6'1"'1'1e 8: Verify that det (BA) = (det B) (ilet A), where A = [: :] .ind B = [!' ~] .
solution:
Let
0

B = [~ ~] and A = [~ ~]
Then.

BA=[~ ~][~ ~]=[kaa+c kb+d

-b]
Therefore, .

I I
det(BA) •= kaa+ c : kb: d = (kab ~ ad) - (kab + be) = ad - .be
det(A) =· I: i:I= ad - be,
1 01 . ,-, .,
= 1~ ., 1 =i' ,.
1. I

. det (B) . , ,. ;)
Then det (B) det (A)= (1) (ad - be)= be: - ad= def (BA). .. ·' d. , . )- = ~~ ~ ~"'"."'77

flie
ff .,

.1._EXERCISE 4.1 l

Compu·te the determinants

. using a cofactor expansion. 2
3 0 4 . 0 5 1 3 -4 3
1. 2 3 2 2. 4 -3 ~ 3. 1
1 2
10 2 4 -4 '-1
5 -1 4· 3 .Q
2 3 -4 5 -2 .4 6
5. 4 O 5 -6. 0 3 -5 . 7· 9j
5 2
5 1 6 ·~ 2 -4 7l 7 3 ·.
-2 5 2 3,
6 .0 0 5 O o·
O 3 0
9. ; 7 2 -5 10. 2 11. 0
0 0 0 -6 -7 5
5 O 4 4 0
8 3 1 8
--·
 11ATBBIIATlc&-II 6 3 2
3 -5
0 -7
4
0 0 2 0 0
-8 14.
9
8
0
-5
-4
6
417 · 0\01
4 . 0 0 0 -6 4
7 3 3 0 0 0 10
7 -1 0 0 13. 5 2 -3 4 2 3
12. 3 0 5 0
2 2 0
2 6 9 -1
8 4 -3 0 0
5
3 12 13 151
Compute the determinants 1 2 -4 2 r
18.
O 5
3 O 4 O 17. 3 1 l ~ 3 4 2 .,
15 2 3 2 16. 4 - 3 1 4 - .
. 0 5 -1 2 4 1 d tenninant of a ma~ In each ~
. ration on the e '
Explore the
effect : an
state[ath~ro]w[co~]a on an [ 3
e~er:n~ ;:: 4] [ 3
:~eds the detennmant.[a
4 J Zl. c d ,
[a + ~
c
+
d
b] , b kd]
19. c
d , ba 20. 5 6 , 5 + 3k 6 + b a c]
[3 2 2 ]

22.[-12 -3! !],[~ : !] ! , : ~- ~

gr~ :rt~1res. ,,
23. [{ ;
2 2 -3 2 ·-

:mure f of ilie
i .) .

27. r~ 1kg]l
O . 28. [~ ,i tJ
29. [~ .· ~ " ~] O l
L~o 1 o o 1 1 - o , o _ _. . Sol
Give reasons for your answers. ·· . '
30. What is the determinant of an elementary scalar matri?c with k on the diagonal?
Verify that det(BA) = (det B) (det A), where.~ i,s the ele~~ntary,ma~ shown and A=[: !]
31.[~ ~] , . 32. [10 . kOJ . . . , 33.
rt k]
to 1 .·
' [3
34., LetA= 4
~]. W~te 5~. Is det (~A) ~ .§ d~t 0(A)? ·· ,

35. LetA=[: •!] and let k be a scalar. Find~ fo~:ii~ ~ ~t-relates d; t (kA) t~ k and det (~)-·

1. 1 20 ~ '
5. -23 1 7. 4 8. -11
,,
'
9. 10 10. -6 . 11. -12
13. 6 14. 9 15. 1 .
12. 36 .. " Ii
16. 2
17. -5 18. 20
19. Two r~s are interchanged. The determinant changes sign; . l r
1
20. One.row times k is added to another row. The determinant does not change. _

21. One row times k is added to an~ther row. The determinant does not chan
22. First row of matrix was multiplie'd with k and resulting detenn;.,.......
. .r._ , • • r . i
ge.
lQJ.t was multiplied with k. 1
-
1

23. Two rows were swapped m given matrix which resulted in multi . - ,
4&........

24. 1 ' - · 25. 1 , P1ymg determinant with -1 .

. . ~ ·._
... 26. k 27 k
~- .. 28. -1 29. -1 ·
I ~ k
.. : ' 32 erifi d 31 verified
.j ! .v e 33. verified - ~- det(SA) ~ 5 det(A) .
.,.. 35. det(kA) • k2 det(A)·
 Determlnanta.W \aw,-41
, pooPEBTIES OF DETERMINANTS

~§::================~~===============
~ 'fhe secret o fd eternunan
. t s li.
e m h ow they change when row operations are performed.
erations on a matrix '
JtoW Op . .
A process on a man:"' ~ t any of its ~tries can pedorm so that all other values on the columns
o[ll.e zero such operation lS a row-operation on the matrix.
t,eC we try to develop as a triangular matrix to a given matrix with the help of row operation.
I-{ere,
.,o,_m,

Example l: By using row operation, co~pute det(A) where

. [1 2 31
A= 0
3
5
7
-4 .
4
Solution:
Here,
1 2 ,. 31
det(A) =
\
~ ; . i ·.
, .
Performing &~RJ - 3R1 then (.

2 1
-4
3 \ .
det(A) = O 5
· \0 1 -5 I
I
J ,. I I I, r

R2
Performing R2➔ 5 then
J' f

2
·. det(A) = s \ 1~ 11s\
:.s
..
' 0 1

Performing&➔& ~ R2 then
I
_.; I - • I

2
det(A)= 5 H 1
0
3
-4/5
-21/5
.

The determinant is tri~gu1ar. So,

1
det (A)= (5)(1) (1) (-21/5) = - 2 .
 MATBBIIATICS-0

s -7 2 2
0 -4
0 3
3
.
Example 2: Compute -S -8 0
0 5 0 -6
goli
Solution: J-Jef
Here
5 -7 2 2
0 3 0 -4
-5 -8 0 3
0 5 0 -6

5 -7 2 2
0 3 0 -4
= 0 C.. performing & ➔ & + R1)
-15 2 5
0 5 0 -6

5 -7 2 2
0 3 0 -=4 . ' &➔ &+SR2 · .
= 0 -15 C. · Performing ~ ➔ 3~ _ SR2 )

I
0 2
0 0 0 2
' . .
= (5) (3) (2) (2)
.'
=60 )

.,'"'.
Th10Nm 4: A . 1
Proof: Suppose a square matrix A has been reduced to an
echelon form U by row operation and row
interchange. H there are r time row interchange then

det(A) = (-lf det(U)

Since U is echelon form so it is biangular, hence *
det (U) 0, figure (i)
det(U) = product of diagonal entries
If A is invertible

, u.JiLo ~ ~ ;1
(::) diagonal elements are pivot elements

(::) det(A) = (-t)' product of pivots:,: 0. fig. (i)

HA is not invertible o o 0:J
det (U) ~ O; figure (ii)
(::) at least one diagonal element is zero
¢:> det(A) = O; fig. (ii)

Therefore, a square matrix A is invertible iff det(A) 0 . .

- A set of vectors {v v
·
} h * · }
t, 2, V3, • • ., Vn W ere V1 E Rn are 1in
' ear Ydependent if
. det [v1 V2 V3 • • • Vn] = 0
and linearly independent if
 det [v1 V2 v
.,........._
~ Iii e#i-41
Use a determinant to d 3. • • • Vn] ~ 0.

Vt =[ ~u [~]
ectde ifv

0 , V2 =:
1,[v;]v~ Ve are[-llin]"e~ly independent or not, when
-7
3 , V3 = 6 , V4 =
3
- -5 2 .
solution:
4 -2
'
ttere,
0 1 2 -1
det[v1 v2 v 3 v,] = ~ 5 -7 3
3 6 2
,. I ' · .~~ -5 . 4 -2
0 1 2 -1
= 2 5 -7 3
0 3 6 2
0 0 -3 - 1
1 2 -1
= -(2) 3 6 ,- ~ 2
0 :..3 1
,· 1 2 :.:·~1
- :;: -2 0. 0 · . ·5 .
·• .Q.. , :-3-.. l [Appl~gR2 ➔ R2-3R1 ]

. =-2 <1> I!. ~I =-2(0+15)=-30 .. o.

This means the given column vectors are linearly·independent.

Example 4: Use deterininants to find out if the maim is Invertible or not,[!

Solution: _. ) 4 t f • ,.

Here, : .
5 0 -1 0 15 9
1 -3 -2 = 1 -3 -2 [Applying R1 ➔ R1 - 5R2]
0 5 3 0 5 3
re (i)
=- (1) 11s
5 3 . 91 ' I•

=(-1) (45 - 45) =0

This means the given matrix is not invertible. .
- Every matrix, whose column vectors are linearly dependent, is not invertible.
Example 5: Use a determinant to decide if vi, v2, V3 are linearly independent where

Vt =[ f], V2 ~ [ !l ~ v, [-}]'

Solution:

Here,
 MATHEMATICS-0
~

-3 2
5
= -7 3 -7
-5 5
~
9
5 -3 2 [ perfumtlng
..:21 · R245R2 + 7R1
= 0 -6 R3~5&-9R
0 72

=
5
o -6
-3
-21
2 [ Performing
R3 4 3R3 + R
J
O O 0
= (5) (-6) (0) ·
=O.
This means the matrix whose ~olumns are Vt, v2 and V3 is not invertible. This means the
column vectors are linearly dependent. so
Note: Sometimes the vector [ f] ~ written as (5, -7, 9).

Numerical importance of determinant calculation byrr:~w·qp_eration 3

A research shows that the evaluation .of an' 11:~ h; ~~ti r~ t . req~es about 2; arithmetic
operations by using rQw ope1c1ti~~ :-:r~ ~~;fore>:'. f
determinant in less than a second becau~~ ?hly·~b~#t 10,0'Q
fi11~p.J;-,f
o~ ptit~r_ can calculate a 25><25
Oh:PPer~~ons are required to compute a
2Sx2S determinant. Whereas, such· detehninanf : computed,fbY,'.:~the computer requires 25! (i.e.
1.Sxl o erations b, cofactor .e~"~~~- ,1,',.}; tt·.
,::::.j:~, _-J1;+J/ 11 .' . µ•.. ,
1
•,I •

Column Operations
Column operations in-an analog<;>u~ JQth~:fOY'f, ,operatio115; we c~_per(orm this operation on
columns of a matrix and a determinant. · ·
Example 6: Evaluate the following by column operations,
1 5 -3
det(A) = ·3 -3 3
2 13 -7
Solution: " -
1 5
-3
Here, det(A) = 3 -3
3
2 13
-7
Performing C2 ➔ C2 - SC1 and C3 ➔ C3 + 3C1 then,
1 0 0 , ..
det(A) = 3 -18 12 I •

2 ~ ,,-.1
r

Performing C3 ➔ C3 + !~ C2 then ,

1 0 0
det(A) = 3 -18 0
2 3 1
This is a triangular matrix. So,
det (A)= (1) (-18)·(1).
a. _ 10
·.t:-e.---

- .
 rem 5:
If A iS n><n matrix then
Determlnanta w \§w:#41
'
rJatriX and Determinants Products
'fhe product oftheorem
and following shows the relationship between tlte determinant of _multiplication of matrices
determinants.

nwc,rem 6• (Multiplicative Property) .

If A and B are n><n matrices then det AB -_ d et A . det .
, 7: S~ow that
Exall\Ple [ det(AB) = (det(A))·(det(B)) h olds f or the matrices
.

A =4
1
3]2 and B =[:
·1]. .
Solution:
Here,
AB =[41 3][6
2 3 2
1]=[3312 107
5 J·
det(AB) = 33
Then, \12 10\
s = 165 - 120 45. =

l\etic
Next, det(A) . det(B) ~ . \4l 3\2 \63 , 1\2 ·= (5).(9) = 45.
• I

5><2.5
•utea Thus, det(AB) = 45 = det(A) • det(B).
! (i.~ Show that det(A + B) :I: det(A) + det(B) holds_for the matrices
Example 8:
A=[! !] ·and u=[~ ~].
ion on
Solution:
Here,
A+ B == 3
'[1 2]4 + _[20 1]3 ·[33 . 3]7 . ·
=

det(A + B) ,:; \; ; \ . == 21 - 9 = 12.

Then,
det(A) + det(B) =
1 2\ + \2o 1\ = (4.-6) + (6-0)· -(-2)
_. .+ (6) -_4.
Next, \3 4 3
det(A + B) :1: det{A) + det(B). _ . ,
Thus, -[1 0] d B = [a bJ . Show that th~ det(A + B) = ·det(A) + det(B) if and only
Example 9: If A- 0 1 an c d
if a+ d = O.
Solution:
1 01\==1
Since, det. A = \ o
det B = c
a b\d = ad - ·be
\
l+a· b ]
Here, A +B= c[ 1+d
1 +a b \ = 1+a + d + ad - be
Then, det(A + B) = \ c 1+d
Suppose, a + d = 0, then d d _be
det (A + B) = 1 +a + +a
:::: 1 +ad-be
 MATBEMATICS-U

Again, det A + det B = 1 + ad - be

Therefore,
det (A) + det (B) = det (A + B)
Conversely,
Suppose, det (A) ~.det,(B) =, det (A + B)
1 + ad - be = 1 + a +· d + ad - be
⇒
⇒ a+d=Q dt~)
al to det(A) + e ·
Caution: If A and 8 are nxn matrix then det(A + 8) may not equ .

Linear Property of Determinant Function . th n det(A) is a linear

H a; be fixed value of a coh1~ of a matrix A. Let T be a transformation e
· function of a;. That is, let the columns of A are "
A = [a1 a2 . . . ai . . . an].
HT is defined by
T(a;) = det(A).
Then, T(aa;) = aT(ai) for any seal~ a.
T(u + v) = T(u).+ T(v).
g h i
Example 10: Find the determinants •of a b c , where
a b .c
d e f = 7•
l
d e f g b., i
Solution:
a b ·.C
,, ...
Let d e f =7
g h i
Now,
g h i

def
a · b c
a b c = (-1) g h i ; ~(-1) (-1)1
def
a
d
g
b
e
h
!I =,(-1)(-1) (7) = 7.

2_HXERCJS~ .4.~ .. . :, .

r
Using properties of determinant, show
, ,
that
~♦ .. •

-3
' - • - • I
·"'' . " l • . • I

I!
5 6 2
1. -3
-1
-ii
: =- ~ -15 -2
8
2. 3
1
-6
5 41
-2 = 2 3 \1 ~53' -22\
1-41 r
6

I
3 . 1 6 3

I~
3 3 -4
3. 0
-4 = ~
-6
-4
5
7
4.
1
0
2 3
5 -4 = 0 1·2 ::l
5 '-4
3 7 4 0 1. -5
Find the determinants by row reduction to echelon form:
1 5
5. -1
-2
-4
-7 ti 6. .
1
3
2 13 -7
5 -3
-3 3
7.
1
-2
3
1
3
-5 7
5 2
-1 2
0 2
4
1
-3
I ,
 Detenmaaat. ID l§w;jafi)
-4

8, I /-3i !!
-7 -5 2
-5
-3 9.
- l
O
-1
3
-1 -3
1 5
2 8
-1 -2
1
0

3
O
4
5
10. -2 -6 2
3
3 -1 O -2
2 -4 -1 -6

7 -3 8 -7
3 9

CoJDbine the methods of row reductio d 3 5 5

2 5 -3 -1 n_~ 2 7
;of;ctor expansion to compute the determinants:
3 0 1 -3 3 O 2 5 4 1
11. -6 0 -4 9 12. 5 4 3 0 4 7 6 2
4 10 -4 -1 4 ~ 6 6 13. 6 -2 -4 0
4 3
-3 -2 1 -4 -6 7 7 O
1 3 0 -3
14. -3 4 -2 8·
3 -4 0 4
a b c
find the determinants of the following, where d e f = 7,
g h i
a b c a b c
a b c
15. d e f 16. ·3d 3e 3£ 17. g h i
Sg Sh Si g h i
d e £
a b c a+d b+e c+f
18. 2d + a 2e + b 2f + c 19. d e f
g h i - g h i
Use determinants to find out if the ma~ is invertible:
0
2 3 OJ
[1 2 1
i_ 7 ~5
[o_ .7
I ~]
20. 1 3 4 21 · 3 8 6 0
s. 4
Use determinants to decide if the set of vectors is linearly independent

22
LU[D[=JJ · 2J. [;Hin1J
25. Determine if the columns ~f A are linearly in?ependent.
24
. [11[t1l=nuJ
,[~ ~ -i]
5 8 0 .
26. Without expanding the determinant, compute
-2 8 -9
-1 7 0
1 1 -4 2

5. 3 6. -18 . 12. 114

11. 120
9. 3 10. 24
15. 35 16. 21
13. 6 14. 0 20. invertible
19. 7
17. -7 18. 14 24. dependent
23. independent
21. not -invertible 22. independent
25. Independent 26. 15
 MATHBMATic&-11 TRANSFORMATIONS
EANff LINEAII
4.3 CRAMER'S RIJLE, VOLIJM .

- det (A(b)) f . = 1 2 .. ., n .
Xi - det(A) or 1 , , t"

. f A is denoted ~y a1, a2, .. ..,_an and the

Proof: Given, A be an invertible nxn matriX. Let the columns O • ,
columns of the nxn identity.matrix I is de~oted by et, e2, ... , en.
Then, Aei = a1, Aei = a2, ..I •ll Aen
., = an.,·. ' . ~I ~-l, r ·. ,
rl1
If, Ax= b for .be Rn. Then, fori = 1, 2, ... , n
A li(X) = A [et
e2 ... X •• • en].
b
= [Aei Ae2 Ax ... Aen]-

,. '= [a1 a2 ... b ... an].

=A (b).
Therefore, by multiplication of determinants,
(det (A)) (det L(x)) = det (A(b)) ~ ;. 'fot i =-1/ 2~ : .. , n. :
'.., .
.i •· •. ' • . 'q, J u
Since I is an identity matrix. So, det (Ii(x)) == Xi .-
Therefore, r
t ij
for i = 1,i ... , n.
I
det (A) Xi = det A(b) .
J t~ - .--1 .~ , .. ' ,. - . ,) I ,'
_ detAi(b)
⇒ Xi - d~t (A) : ; fori F 1, 2, :···! n.

Since A ts inv~rtible. So,·det(A) ~ 0. This me~' Xi has u¼q~e solution.

NumericalN
If A is a 3x3,- ~b'ii-ivith c
arithmetic is \Joo,r<iered ah
order tlien the ctamer:_s· , ..
compu~tional PQ.fut olW.
Example 11: By using Cramer's rule, solve the system of equations
3x1-2x2= 6
-Sx1"+ 4x2 = 8. . '
Solution:
Taking the given system as in Ax = b and choosing it as,

Here,
A=[! - ~
n A,(b) =[! iJ, A,(b) =[!s :J.
 3 -2,4 = 12 - 10 == 2 ~ 0. Detenntnanta w l§rii-1
det(A) = -5

'
So, the system has unique solution and the )Jrocess is possible.
Therefore, by Cramer's rule,
_ det A1 (!?} _ 24 + 16 40
Xi - det(A) - 2 - == 2 = 20 .
2 _ det A2 (!?} _ 24 + 30 54
x - det {A) - 2 ==2== 27 ·
Thus, X1 = 20, X2 = 27 be the solution of given system.

Example 12: solution.

Using Cramer rule delennined the value of s for which the system has unique
3sx1 -2x2 =4
- 6x1 + SX2 = 1.
Solution:

Taking the given system as in the form Ax == b then

3s
A = [ -6 -2J
s , X = [X17
xJ and b == [4]
1
.
Then,

Therefore,
A,(b) =[: ~]
2
and A,(b) =[~ iJ
det A1{b) = 4s + 2 = 2{2s + 1)
and,det A2{b) = 3s + 2~ :" 3(~ +_~) , , · . . · ·
also,
det{A) = 3s2 - 12
= 31s2 _·4 = 3(s - 2) (s + 2).
Now by Cramer's rule,
det A1 (b) '2(2s + l)
x1 - det {A) - 3{~ - 2){s + 2)
det A2(b) s· + 8
x2 - det {A) - {s - 2){s + 2)

Hence, system has uru·que solution when s :;:. 2 and s :;:. - ~-· ...
. -1
A Formula for A
"'IP> , ',
Thep,em 8: An lnvirse Form~la ..
.
LetAbeanihvertiblenxn ~ ' trix- Then
.
- ·• -1 • ~ Aaj(A) ..
A =det(A ·
~ then det(A) 'I:- 0• So A-1 exists.
Proof: Let A is invertible
. A wehave •
Since, for any square matrtx ' . A= det(A) I
A. AdJ'(A) = AdJ(A) . ) _
1 Ad. A) A-I
A l Adj(Al) = (det(A) J( .
⇒ (;
· det(A) .
 MATBEMATICS-ll I

~ A-1 = de:(A) Adj(A) .

.
Example 13: Find the inverse of the matrtx 2
·[13. 5014] . 1 1

I! ~ !I
Solution: Here,

31 ~ ~ I-sI~ iI 1~ ~ I 4
det(A) = = +
=-3 +5 +4.
=6.
So, the inverse of A exists.
The cofactors of A are,
Cn = (-1)2 I~ ii =-l.
C12 = (-1)2 I~ i\ = 1. I •• •

C13 = (-1)4 1i ~I =1· Cn =(-1)3 \,~1 i\= -:1.

C22 = (-1)4 1 ~ ii = -5.
€23 = (~1)5 \ ~ ~ \ = 7.

~ 1 =<-1>·1~ i\= 5. -_
= (-1) ~ i\-=_1.
C32
5
\

..
I '• / \_ ~

I~ ~I =-5.
I ,

c~=<-1>6
Then, { '. - -1 -1 · '5
Adj(A) = Transpose of iru::~ of co1acfu o'f A,= Li '~--~1··
Now, using the Theorem-8, i.e. inverse form[_'t' -1 ~51·]-.
A-t = l Adj(A) =!6 1 - -5
· det (A) l 7 -5

[)etenninants as Area or Volume

The geometrical interpretation of 2x2 determinants is area of parallelogram and 3x3 d e t ~ t is
volume of a parallelopiped.
The following theorem concise the concept.
is2X2 JllatriX, tlie area of n ~p ogr · e
. ·•-~. co iimns-of Ais
. 1:e ,t' ·~ . . • . ·• oftti~~,
. ;
(A).
· · i ·: mnnsofA.
f det A.

Example 14: Find the area of the parallelogram determined by column of matrix

A=[~ !]
Solution: Given that the parallelogram determin~ by the column of the matrix
 Determlnanta Ill i§w§◄
- A=[23 5]6 ·
. ,uHere, - t 1, •

- det (A) = 12 .. s 1 . .,
• •( ., l 3 6 =(12) - (15) =- 3.
Then, the area of the parallelogram is '

• ~ area • I det (A) I = I-3 I = 3.

Exafflple 15: Fmd the area of the Patiillelo ·' • .
Solution: gram whose vertic~s are (O, -2), (6, -1), (-3,-1), (3, 2).
Given vertices of a parillelogram are (0 -2) ·
vertex becomes at origin as; ' ' (6'-l), (-3, l), (3, 2). Translate the vertices so as one
(0, -2) + (0, 2) = (0, 0), (6, -1) + (0, 2) = (6, 1),
(-3, l) +_ (O, 2) = (-3, 3), (3, 2) + (0, 2) = (3, 4). ··
b ethlograml IS shifted with vertices (0, 0), (6, 1), (-3, 3), (3, 4). So, the parallelogram is
parall
Thdentermillth~ed
e y eco umnsof

A= 1 [6 -3J 3 ·
Then,
6
det(A) = 1l _3 =18 +_3 =21 .
-31 ,,
., Thµs, the~ of~~ p~@,ilelogram is, . .'
area of A = I det(A) I = I 21 I = 21.
.,
Example 16: Find the volume of the parallelepiped with one vertex at origin and the adjacent
vertices at (1, 4, 0), (-2, -5, 2) and (-1, 2, -1).
Solution:
Since the one vertex of the p~allelepiped is ~t orisi!t and the adjacent vertices are at (1, 4,, 0), (-
2, -5, 2) and (-1,·2, -1). Then, · ' ' ' , . ·
1 ~2 ~i .,.<.. I ,· I ! ' 1r U ' \ l

det (A) = 4 -5 2 .
0 2 -1 ·,

=1 I1 _; I·'+~ I~ 2 1 1~ , -S12 ··
-1 -l O
I'

=1-8-8 =-15.
Thus, the volume of the parallelepiped. is, ._
volume of A= Idet(A) I = 1-15 1= 15·
 MATHBMATICS-D " . d if Sis a parall
3><3 Ill8.trtX A an
Likewise, if T: R3 ➔ R3 be determined by .
bt = [ 1] and ~ = [~ ;' ~d
9
volume of T s = det A · volume of
•
. ed by vectors 3 1et
Example 17: Let S be a parallelogram deternun •
- • 1• under mappmg x➔Ax.
A = [~ -: ] • Compute the area of tmage of S 1•1

Solution: bt _ [1.J•~d 1n = [il So,

<;;iven that S is the parallelogram d~termined _b y ~e vect(?rs,• - · 3 · ·

~et(S) = I! ~I = 1-15 = -14. , .

Thus,
Area of S = 1-14 I = 14.
And given that ·

[1 ·-21]
.r
- •r. . A= 0
Then,

det(A) = 1
1O -112 = 2.
Therefore, the a area of Sunder the mapping x➔Ax is,
area of image of S = Area of T(S) 1
_ • • •

= Idet A I [Area of S] = 2x14 = 28 sq. urut

Example 18: Let a and b are positive numbers. Find the , area of the region E- bouiufed I by the
. . Xt2 -z 1
e llipse w h ose equation 1s a2 + b2 = .
. I ' -. xr
'

. ii . . . -, ..
'l
Solution:
Let,
· Let: · .:s u= [:j .r ·· fml i =[;j :, : · : ,. ·,,,fa,cc: ,: ·> •,t-, -~t ,. ~ .n ·

Let E be the image of unit disk D under a linear tr~fc:>rmation T determined by matrix A with
Au =x. Then, ., .
Ut =a,
X1
U2
X2
=1, \. ..
Since u1, u2 lies in the unit disk
I
with ut2 ti •·u22~• 1 Jf and only-il
. •
x is in E with
X1 2 Xr ' .
-+-<1
a2 1,2-
Then, area of ellipse = area of T(D)
. . .'
= I det (A) I {area of of ,I

= ab.1t [": D is an unit disk]

= 1t ab.

· ·· 1 ·o ro 1] -.
Example 19: Let the_four vertices 0(0, 0), A(t, 0), B(0, 1) and C(l, 1) of a unit squar be ·
e
· td
represen e
by 2>C4 mabix: Lo O 1 1 . Investigate and interpret geometri.c~y efftd of the
pre-multiplication of this matrix tiy the 2><2 matnx[42. -1]
l ·
Solution:
The matrix represented to a square having vertices at.O(0 0) A( .
· , , 1, 0), B(0, 1) and C(:
 . Determinants w §mR41
; s =[00 1
0
o 1]
1 1 r
So, . · det(S) = I~ 1 I0 == 1
and given matrix is

A=[~- ~1]
So, det(A) = j 4
r 2
-11
1 ==4+2=6
. .
Therefore, the effect of re- . . ·
p 4 multiplication of S by A .
S' =AS=[·2 -lJ[O 1 0 is,
1 o to 1 1] = [o 4 -1 3]
and 1 0 2 1- 3

area of 5 = ri 4 -111
1- ' '
1

2 1 4
This means the ve · = I ( ) - (-2) I = I6 I =6.
' rtices of the effect f ' .
C (3, 3). . o llie square A are 0'(0, 0)' A'(4, 2) , B;(-1, 1) and

q n. .._ B'
Lr-~ A'
A
•· r l 1 • fJ ,,

le figure I figure 2
Here, (area of S)(area of ~ ) .1 . . I. ,

rh~ -.; n = I det (S) 11 cl~t (A) 1

.l
. , .. : ='II~ ' ~11"1,~- Lill= 111161 ·=6=AreaofS'.
This
by me~,the transfoimation mat:NcJl\i · th .
the scalar times of the ma~. . . - giv:..es_ at th~ ar~a 9f1th~ giyen unit square is increased

..1 HXERCISE 4.3

A. Use Cramer's rule to compute the solutions of the systems.
1. Sx1 + 7-x2 = 3 - · . 2~ 4x1 + x2 ~ 6 3. 3x1 - 2x2 = 7
2x1 + 4x2 = 1 ) sx1 + 2x2 = 1 - Sx1 + 6x2 ·= -5
'
4. 5. 2xt + X2 = 7 6. 2x1 + X3 = 4
+ X2
- Sx1 +. 3x2 = 9
. ·-3x1 + X3 ='-- -8 .-x1 + 2X3 = 2
3x1 - X2 = -5
. r x2 + 2x3 = -3 3x1 + x2 + 3X3 = -2

B. Dete~e the values of the .parameter s for which the system has a unique solution, an~
describe the·solution. .
7- . ,6sx1 + ~ 2 = 5 8. 3sx1 - Sx2 = 3
9x1 t 2sx = -2 9x1 + Ssx2 = 2
2
9. sx1 - 2sx2 = -1 10. 2sx1 + x2 = l
3x1 + 6sx = 4 3sx1 + 6sx2 = 2
2
C. Compute the adjoi!tt (i.e. adjugate) of the given matrix, and then find inver"'; Ill the matrix.
.-
 MATHEJIATJCS-D

~
13. [!
11
· [l 11] 12 [ !12 io]

rn
14 [~
. 2 3
~ ~
~J -1 1
-2 3
15. [3 0 ~2] 16.

1 . ts if A is invertible.
17. Show that if A is 2 x 2, then A- eXJs _ . why all the entries in A-1 are integers_
18. Suppa;e that all the entries in A are integers and det A - l. Explain .
Ii ted
Find the area of the parallelogram whose vertices are 8 •
3 2
19. (0, 0), (5, 2), (6, 4), (11, 6) 20. (0, 0), (-1, 3), (4, - 5), ( , - )
21. (-1, 0)_, (0, 5), (1, -4), (2, 1) 22. (0, -2), (6, -1), (-3, 1), (3, 2)
1
23. If A is invertible then shows that det(A-1) = det(A) ·
· • · and adjucent vertices are
·gm
24. Find the volume of paraJ)elopiped with one vertex is on
(1, 0, -2), (1, 2, 4) and (7, 1, 4).

,. 25. (i) Let S be the paralleioii:am d e t e ~. by the__;':"tors, b, =

A = _3[6 -2J Compute the area of the· image of s_.~der the mapping
2 .
·
[-;J, b:z = [ f]
x➔
Ax.
and let

(ii) Repeat (i) with bi = [_~], ~ = [~] ~d let A = [l !i] ~

26. Investigate and mt:erpret geo~ etrically the _transfo~ation pf the unit[{u~ :]hose vertices are

0(0, 0, 0), A(l, 0, ~), B(q, 1, 1),and C(l, 1; 1) effected by lhe-3><3 matrix 01- 1· b '. -r
· · . 0- 0 1 1111111111111

1· 5/6]
[ -1/6 2· 5/3]
[ -2/3 3. 4.

1.5]
5.
[ 4
. -3.5 6. rn1
3s+2 2s-9
8. s e R x1 s(~ + 3) , x2 ~-S(s2 + 3)

! _ 2(3s -1) , _ 1
10. S ~ 0, 4, Xt - 3s(4s - l)' X2 - 3(4s - l)

13.¾
19.8
ri -1
-5
7 _!] 14. ~2

20. 7
r-s
3
2
-3 !]
21. 14
24. 22 cubic units 25. (i) 24; (ii) 20

□□□
 .I I

CljllPffR

j let

VECTOR SPACES
· are - . .

LEARNING OUTCOMES

4ller studyfng tbfs cbapter you sbould be~ to:

~ll. Vector Space and Subspace

Null Space and Column Space, an
d Unear Transformation
~ Unearly Independent Sets, Bases
~ Coordinate System
 MATHEMATICS-II

INTRODUCTION . .
- ality linear algebra 1s the study of Vert-~
tor space In re '
This section introduce the concept Of vec
-"'?
· f rmations) They form the fundarn.-.
ear trans o · ~•ltai
space$ ..and the functiofls of vector spaces (lin . . g course of algebra. One we define
objects which we will be studying throughout the rematrun Their importance lies in the fact n,_a.
• d ...., rti of vector spaces. ...~t
vector space. We will stu u1.,e prope es . bout vector spaces. Thus each fact n,_t
. . be hr sed as a question a ... ~
many mathematical questions c~ rep a d" information about many different
we can prove· about vectorfspaces gives us correspon mg
mathe~tical questions . . --:· · ~~ -
., I' •. • • #t ..

--
..~

.
j ,. ... .,.

s.1--vECJDR 1SPtl~ itND SIJBSPACE

- -
,,

TbenF
Examples:
1: The set of real numbers is a field.
2: The set of rational numbers is a field.
3: The set of complex numbers is a field. _
4: The set of integers is not a field. Since there is no multiplicative inverse of every non-zero
1
integers. In particular 3 e Z but 3 e Z.

r•1i~~~:t~:~~~'¥~--
-_. . . or all -ve y, there exists -ve ··,

•, -,~;-Vr~0 ·:titJJ!tiJriit U~:~;fb.f>~---~xP 'ft 1

. ;:._)· -ni,uJtipijcation by ~alar _\

-J~-~(~~) _ --~or·all a, be K;hv&¥;dt1ibn ,.;~ ~~:-.
i)J a ~ l?)v:~~vi~lw·:- .,.-i.)r;;i,:,:-. r.mu~--j bii:ti .~• HiHl..•'l·•·"'-~ -"""
i}?~}.~(~f +,y2):==,av1 + av2, :for~~ttt/ v2,s V a~·]?-e I(
• ·,a . · ... .. tip · 'lic,ative:ideritity: le K such that
._.~,,,.,...,_ ;z~-,~/~~t~ ·",}.'!t.:.'.·~~-,~l, .., ?\-_'·. _J.·:/ -,/_•/<·
:~i:t. . . ~'u
l t(l:I,« . I

C<.,•I,,,.--,...,·,,/,~it1'-1f?~~-t!f~v it::-~-- for all v e V.

 Vector 8pacej W §iijael :
In other words, a ROn-<!JJtpty set V is called a Vector apace over· thedield K. H V forms an
additive abelian group and scalar lnultipliC'1tion of any elements of V by any element of K, and
sum of two elements of V is •gain in V and satisfied the condition (I), (g)' and (Ii). .
E,<aD'Ple 1: &ample 1: Let V • {(: ~, a, b, ~,de a}
be the set of all 2•2 matrices with real
enbies. Then y ~ a vector space over 'the field R. · ·

~
ExaJDple 2:
Let n 0, lJ!e,n
t!ie. oet P. of, polynomlaJs of degree .at most !1 consists . of all
polynomials of the form. . .
P(t) = ao+ a1t + ••• + antn ...... (i)
where, the coefficient cl(), •• • , an and the Variable tare real numbers.

The degree of P is the highest power of t in (i) whose coefficient i~ not zero. If P(t) = ao 0, the *
degree of P is zero. If all the coefficients are zerb, P is called the zero polynomial and the
additive identity. If 4(t) =ho+ ~t + ... + bntn,_then the su1;11 ~ + q is defined b!
(p + q)t = p(t) + q(t) I
I I

= (ao + bo) + (~1 + .b i)t + ... :t @n + bn) tn·

and the scalar multiple cp(t) is the polynomial defined by
• I

(cp)(t) =cp(t) =cao + (ca1)(t) + ·:.. + (can) tn_

Oearl the zero polynomial~ a zero vector and.(-1) pis ilie negative of P. Thus, Pn is a vector
space.y . .

Example 3: The,set of recu

'- 1 n. um
. bers.foim 'a vector space'over
. itself.
. -, _. - b
Example 4: The set of all mxn ~a~ce~ fo~s a vector,sr a~: ~v~r the ,set of ~eal num ers.
'V = f(a11): a11 e R}, . .:·. 1·11 ·n ..1-:, .
a set of all mxn matrices is a vector space ov:~! ~ . , ,., i

Example 5: Let F be the set_of all real v alue djunctions

. such that

(i) (f + g) (x) = f(x) + g(x) . .i· • ,i:· . _j •· ll• . J , ~. • .. _ ·

e' .. , . . . (ii)(af) (z) = a£(%) for alb, a e R. · :· , ., . ,

- Then Fis a vector space over Ji..'. .. ,. . :, I f . , . (3s ,2 + 5s) i~ n~t a v~cto~ space ~y '
"$ "nR2ofthe orm
Show that the ~et H ~ all,p01,
< • • ·t
Example.~: I er ~:1~.. ffiultipiicatiqn. ~inil, the. sp~cific vec or u
' . . th!lt it is. n.,t do~e~ ~ 4 . ~ . , ,
shq~g ·r- . . h that cu is not m H. . :
in H and scalar c sue . .l
. r •r ,,.
Solution: · , ·. ,. '' : ~( 3 ) ( 6 ) .
. .'
Suppose u -
~ '( 3) e H and c = 2 then. cu = ' 7 = 14
· ·
- . '., ;
, ·,
rf
J
7 ♦ t ..,t..
o
•
I I '•

Is cu eH? foithis '

• ' • ,

C~sJ = c:)
3s= 6 ⇒ s= 2 and2+5s=14 ⇒ s- 5
_ gwhich is impossible

Thus, 2u ~ H.
 11ATBSIIATIC&-II

i, ax,, ......, axn) e R is a vector spa

•• , Xn) , y = (yi, y2, ..•, yn) e V, a

Example 7: Compute u + v, u - 2v and 2u + v where u - ' [1] [1]

- -3 v= -1
.3
2
I

Solution:

,.
In other word, let V be a vector space over the field K. Then a non-empty subset W of V is called
a subspace of V if W satisfies the conditions:
(i) Wt + W2 E W for all Wt, Wi E w.
(ii) awe W for all we W, a'e K.
(iii) 0 e W.
Moreover, the single equivalent condition with the above three condition of a .vector subspace is
aw1 + bw2 e W for all a, b E: K and Wt, w2 e W. For a = b = 1, the first condition holds, if b = Othen
the second condition holds. Similarly if a = b = 0, then third condition holds. ·
' .
Example 8: The set consisting of only the zero vector in ·a vector space V is a subspace of V,
called the zero subspace and written as (0). It is also called trivial subspace of V.

Example 9: Let V = R= 3
{~J x, y, z e R} is a vector space over R and the set W •

{lJ s, tare real} is a subspace of V, then show that w is. a subspace of ·R3.

Solution:

(i) Taking, 0 = [i] e W, is an zero element in W.

 (ii) For all a, f3 Rand w
E 1 == [::J0 , W2 ,.
[82]
t2 e W th
o en

aw1 + fJAw2 •[as1 + P82J ·... '

Clt1 + Pt2 w I ' ;
0 e .
Hence, W is a subspace of V.
example 10: The vector· space R2 .
not a subs IS
"] , pace of R3 beca
Exall1Ple 11: Let W = [ : x ~ 0, > l . . use R2 is not subset of RJ.
y y _ OJ prove that .
not closed under sea) . W is not subspace of 2
Solution ar illultiplicatio~. R by showing that it is

Since u =[;] e Wand c = -1 then

cu= -3-~J ~w
. [ '~ I•' 1:"I·• ... '
, .,.
-

This means Wis not subspace of R2.

Example 12: Let~~ [ !Jt1 ~, a}

.c e_ ,is av~~• _S pace~ the field R. Then

. ~zrrrr~;~;~r .' ' '·, "

is not a subspace of v.
Solution:

. UWi + f3w2,= [ ti; t2S2]

S1 +
1
E W.
' ..
Therefore, Wis not a subspace-of V.fJternatively
I - 1r- •f
O IS not the
I member
' of the• W. Hence Wis not
a subspace of V. , · 1 ,-, , '

Example 13: A plane in Rl not through the origin is not a subspace of R3, because the plane does
not contaht :the ze~ vector of R 3• Similarly -~ line in R 2'not through the origin is not
a subspace of R 2•
QaJnple 14: Let v 1 and v2 in a vector space V. Defille .
I

ff • Span fVt, v2) = fav1 + Pv~ CX,.pE . RJ ,

then H is a subspace of V.
Solution: l .
Taking a = f3 = 0 then Oe H. . · :1
And, taking a, f3 e K then for all w1. w2 e H with
 MATHEMATICS-D

w1 = a1v1 + P1v2
w2= a2v1 + P2v2
Then A - (r,r, + Aa2)v1 + (aP1 + PP2)v2
aw1 + pW2 - """"1 p
= <XJVt + p3v2 e H.
where, aa1 + Pa2 =<XJ, aP1 + PP2 =P3e K
Therefore, His a subspace of v.

More precisely if for any w e W there exists ai e K such that

.m
W = a1V1 + a2V2 + ... + anVn. ' }
Then the subspace Wis spannedby v1.V2 .. • Vn and W e write W =, Span
, {v1,
, V2, ••• , Vn •
J,I • ,-

Note: For y e subspace of Rn spanned by v1, v2 ..., v p

~ye span {v1, v2, ..., vp}

I
,.
~ y = X1V1 + X2V2 + ... + XpVp

~ Augmentea ~atrix fvt ·v; 'v3 . ... ·' Vp ~ y] co¥tsistent _

Example 15: Let W = span {vi, .. .-, vp}, whre vi, ... , vp are in a vector space V. Show that V1 is in W
for 1 SkSp.
Solution:
Since, v1 =lv1 + Ov2 + ... + Ovp • '.Lr··

Thus, v1 is written as linear combination of v1, v2, .... Vp vectors

Thus, Vt span W. Hence Vt e W
~, ' . ... - "1

In general Vk..; Ovt + Ov2 + ... + ·1vk + Ovk+t + .~. + Ovp

Hence, Vk e W for 1 S k Sp.

Example 16: For what values of h will y be in the suJ;space of R 3 spanned by vi, v2, v if
3

1 ;. •
~•{hl,v,·=[~Jv,=[!}ndy~[1]. I I ,

' '
S9. ution; , ~ . .
,. Le!.y l>e in a subsRace of R 3 spanned by v1, v2 and v3 then it. is p.ossible to find, -~ · d R
such that ' '' · a, ..,, an 'Y e
·y = <XV1 + ~V2 + "(V3

ffJ={U +~GJ tJ!l

. ,, 'I

i.e. the system Ax = b, A = [-~

-2
!
_7
1]
0
[-4]
'b = h3 is consistent.
1

• T• •

, .,
'.J
I I
 Vector Spaces W §§a 51
. .•

j >'fl J , • .1

Given that the given system is consistent. Thi .

s me-ans.we should have
h-S=Q ·
J ., ..._ct I -
f =~H h=S~ • i .
. ,•
Theorem 1

I • •

Example
17= Let H - f(a - 3J>, b - ii, a, b): aand_b ;in R}. Sltow.that His~ subspace .of R4.
Solution 1

tW
f ·- •

J ' I
·' .
This shows that H = Span {v1, v2}, where v1 and v2 are the vectors from R4. Thus H is a subspace of R4
by theorem 1. i'-

EXER(~ISI~ 5.1
I
1. Let Wbe the first quadrant. in the xy-plane,
. '
i.e.
, . . J' Inf. . '
Letw=[;];x?;O,y?;O}. IJ •

a. Hu and v are in WI is u + v in'W? Why? I

0

b· • wand specific scalar c such that cu is not W.

p·md a specifi'c vector u m

2· . ' Let W be the.union of the first ~ d third qtfudrants in the xy-plarte i.e:, W = {[;] x y?; o}
' . . J '
:i I
.~
• •
·, .
a. H u is in W and c is any scalar, is cu in W? Why?
 11ATBBIIATIC8-II
· th tu+ vis not in Wand hence-Wis not .
b. Find specific vectors u and v in W such a a\>~
space.
th unit circle in the xy-plane. That .
3. Let H be the set of points inside and on e IS, \~
· ectOrs or a vector and a scalar to sh°"'
H = [ ; ] : x2 + y2 s 1}. Find a specific example, two v . • ~t

His not a subspace of R2•

· riate value of n. Justify your answ
4. Determine if the given set is a subspace of Pn for an approp ers
a. All polynomials of the form P(t) = at2, where a is in R
b. All polynomials of the form ~(t) = a + t2, where a is in R:
c. All polynomials of degree at most 3, with integers as coefficients.

d. All polynomials in Pn such that P(O) = 0 .

5. Let H be the set of all vectors of the form [-s~] .

~
F~d a vector v in R3 such that H = span {v\.
.

, Why does this show that H is a subspace of R3? I I

I • • I ;

I 6.

7.
Let H be the set of all vectors of the fonri
-
Let W be the set of all vectors of the form [
r~~] .
L-t
5b+2c] ·
~
Show that H 'i s a slilfspace ofR3•

·
where b and c •~
' r ·
ar~itrary.-Sho': that W bi
·

a subspace of R3•

8. ls W be tlte sd of all vecrors of .tlte fo~[:t::] :.~ow-~t ~ -IB ~ :f of~•-

9. • ~IV•=LUv,=[Hv,=mandw~m
a. Is win {v11 v2, V3} ? How many vectors are in {v1, v2, v3}? · ·
b. Ho~ many vectors are in span {v1, v2, v3}?
' . • I •

c. Is win the subspace spanned by {v1, v2, V3}? Why?

10. ~t Vt =LU =[!], V2 V3 m ~ =[t].

= and W b W is fue su~pace ~ by \vu v~ v,}1

Why? • : I I,< f

11. Let W ~ the set of all vectors o,f the form give,n below, where a, band c represent arbitrary real
a
numbers. In each case, either find set s of vectors that spans w or ; ve an example to shoW
that W is not a vector space.
r •
 VectDr Bpwe ta ICaAffla!I
",·
, l
t
a.
[3a+b]
!
a 5b : a, b real
} b. 1-a+l] }
a- 6b : a, be R
2b+a
C. r~-b1
~~ b.,c 1 : a, real
~

¾t
d,
[~+3b]
•;~~c :a,beR
} {~la+ b+c 2}
e. = £. {~l •-3b-c = o}
~~

1. (a) u +of
the sum
Thus, v .enon-negative
W numbers are non-neg• tive so, u + v has non-negative entries.
.

n {v}. 2. (a) cu= c[;] =[~]is in W since_xy ~.o,·~ (ex) (cy) = ~2 (xy) ~ 0
I

(b) usrneWandv•[=~Jew; u+v•[~2],. ~

3. 4 [~:~] = [~] E H_
' (c) No . (d) yes
(b) No
4. (a) yes
Wis
5. v=[t] I

. ' -
(a) No, 3
(b) Infinitely many vectors (c) yes
9.
10. yes
11. (a) Set does not contains the zero vector, so it- .;.s.not.vector
- space
.. ..

(b) not contains zero vector . 1 : 1

.
:J .

~}[0.;[~l1] span W so Wis : ector sp~ e

' I ., . ' •' I 10
r
l

(e) Set does not contain zero vector so it is not vector space.
'I I

(I) { [!o} U] span W so W is vector space

 MATJIEMATic&-11 All
1JMN SPACE. AND LINE - I

S.2 NIJLL SPACE AND COL , . 1

TRANSFORMATION

Definition (Null pace)

Let A be a m:1Cn matrix then null~pace1of the pl
NulA = {x: x e,R.n: Ax= O}.
eous ~uation Ax = 0.
Nul A is the set of all solutions t~ the homogen · · . [ 5]
' [ 1 -3 -2]· d t mu·ne if u = 3 belongs to the n,,11
Example 1: Example 1: Let A = -5 , 9 l
then e e . . . ,-.2 '"'ll

1 , . . r • • •. , " ' (l I

space of A.
Solution:
Here,
.,

· 1bis means, u is in Nul A.

r
,
y the set
inn

Oearly, Nul A is non-empty since A.O= 0. Since ~ be mxn matrix so Oe Nul A. ~oreover, let u
and v e Nul A, then
Au= 0, Av= 0.
For every a, f3 e K we have
A{u + v) =Au+ Av= 0.
A(au + f3v)' =\ f A(u) -?pA(t ) ',; '() ., ,. .
So, au + f3v e Nul A
Therefore, Nul A is a subspace of Rn.
.'
Example 2: Find the spanning set of.the null spa_~ of the matrix.·
-3 6 -1 1 -7] I '

A = [ 1 -2 2 3 -1 . Also find null A. ...:

2 -4 5 8 -4
l '•

Solution: • f

I \ I I
1
To find the spanning set for the null space of the matrix A we have to
501
• _ o
and find the set of vectors such that ve the equation Ax -

x = y1u + y2v + y3w

where x e Rs, y1, y2, y3 e K.
Now, Ax=O
 Vector 8pacea Ill t§w:§51
-3 6 -1 1 -7 : OJ
- [ 1 -2 2 3 -1 : O I
2 -4 5 8 -4 :0
.Applying R2 H R1 we have
•
1 :...2 2
- [ -3 6 -1 3 -1
1
: OJ
-7 :- 0
. 2 -4 5 8 -4 : 0
Applying R2 ➔ ~ + 3R1 and RJ ➔ RJ _ 2R1 then
1 -2 2 3 -1 . OJ
- [00 00 15 ·210 -10 ; o
-2 : o

- [~ 0 ~ ~ =! :~l
2

0 0 0 0 0 : ~J
Here, x2, X4 and Xs are free variable.
.~ [10 -2
0 1
0 -12 -23 ·; OJ0
0 0 0 0 0 :0
So,
Xt - 2x2 - X4 + 3Xs = 0
X2 is free
7
"3 + 2x.t - 2xs = 0

X4 is free
Xs is free
.
1.e. x1 = 2x2 +X4 - 3xs
X2 is free
"3 = -2'4 + 2xs
X4 is free
xs is free
1 . -3
Xt 2x2 + X4-3Xs 0 0
X2 X2 -2- + Xs , 2
-2x4 -t!-2.xs = X2
X3 = 1 0
X4 X4 0 1
X Xs I •

Here, Nul A -= {x : x e R5, Ax = O}

 11ATBSIIATIC8-II
1 -3
2
1 0
0 ...
2 'l • • •
0 -2 +Xs

1 0
0
0 1
O . as 8 linear combination of u, v, w. Hence, SJ>at\ning
So, every element of Nul A can be expressed . ·
set of Nul A is {u, v, w}. 'ft

Example 3: Find a matrix A such that W = Col A where

r,
Solution:

I Here, W={{t] +b[J}~ be~}.

=span{[!].[!]}.
Thus, the matrix A=[:-7 ~lJ.
0

Proof: Let, .

Col A= {be _Rm: b = Af for some xe Rn}.

• I •
Here A: Rn ➔ Rm then ColA Lis nbn~mpty
- f
because
:
I
for Oe• Rm there is Oe Rn such that
.. I

A.O = 0. Hence, 0 e ColA. · • -- , •

I •

For any~~ e ColA then there exi_sts,x1, X2 e Rn such that

I ,-

For any a and f3 e R we have

abt + 13~ = a Ax1 + 13 Ax2

= A(ax1 + f3x2).
 Veato,- 1lp a■■ ID \§w,na !I
5iJlce, abt + Pl>2 e Rm, OXt + lh2 e Rn. So,

A(ax1 + Px2) • abi + Ph:z e Col A.

1'hi.s means, Col A is a subspace of Rm.
~ate· The column space of an m >c n matrix A 18
.
1, • • all of Rm if and only if equation Ax • b has a
solution for each b m Rm.

ftlE COfflWAST' BE'IWEEN NUL A AND Col A

- NulA
-1. Nul A is a subspace of Rn. 1.
Col A
Col A is subspace of Rm.
-2. There is no obvious is relation between 2. There is an obvious relation between Col
Nul A and the entries in A.
- ,
.
A and the enbiese in A, since each
- - ... column of A is in Col A.

3. A typical vector v in Nu.I A has the 3. A typical vector v in Col A has the
property that Av= 0. - ,. ,. : ptopeity ·that , the· •equation Ax = v is
co~is,tent.

4. Nu.I A = I OI if and only if the equation 4. Col A = Rm if and only if the equation
Ax = Ohas only the trivial solution. Ax = b has a solution for every b in Rn
onto Rm. -

Example 4:
4
-5
7
! HI
>(a): .; If the column space of A is a sub-space.of Ilk, k =?
f (b)' ,;If th~ riull spa~e of ;,._ is a ·subspace of Rk, k·=? . .

Solution: . . -· . -f • · , J. i ·~ - • · f ~· k- 3
.. . . .. ,~ ,, . . thr so Col A is a sub-space o R i.e. -: . ,
Since the column of A has entries ee
· of
b-s ce R 4 so k = 4.
Similarly, the i::tull space of A ts su pa
.

2 4 -2 1] ,. · ColA and NulA. ·

7 3 find a nonzero vector _
m ,
Example 5: LetA= [ - 2 -5
3 7 -8 6 . '
Solution:

s, I.
 JIATHEJIATICS-11 "f.. . ,( r t •

ForNulA, ., - . (~
n ..

,r
:nu• '
H~ X3 free Varli!bles~So
Xt = - 9XJ, X2 = 5XJ. x.t= Q

Thaefure, x= [ f] ~m = so
LetXJ = 1.

Thus, x = (-9, 5, 1,[0: ~ n:n-ze:•v:]or in N[r]· . [.3 ] . 1

••
. -.'
E
, xam pie 6: Let A = -2 -5 7' 3. , u = -1 and v = -l3 ·
3 7 -8 6. b· .
, f 1, 1'
I . [• .
, (a) Determine if u is in NµIA. Could u be in Col A?

f
I Solution:
(b) Determine if Vis iii ColA._Could V be in Nul ~?
t -- j ••• - , ,, . '
(a) If u is in Nul A when' Au = 0
I• '•' }
Here,

I -• [
,. I

Obviously, u is not a solution-of ~ O.!JSo· ti ~is-no'£ ii\ Nul .JA.. 'Since -t::(-is 'four <!iml!I'tsiQnal vector.
So u can not be the member of column s~~ce :'Xltj~~-is {tle, s~~~ac~ 9£ ~ -ee~dim~ional vector
spaces. · ·
(b) To confirm the vector vis in Col A it is sufficient to show that the system of linear eq~ation
Ax = vis consistent. If ' ' · · ~ · ·· ,. -, · " 1
, • • : • • •• • , ,

2 4 J - •
[A v] = [ ; -5
7
By applying elemen~ row operation '.

-01
[
2 4 -2
-5
1
-4 :-2 .
:3] ":"

.r. i: .:~
0 0 0 17 : 1
At this point it is clear that the equation Ax = v of consistent, so v is in col._A._~Slll_ce Nul A j.s a
subspace of ~4 so the vector v with only three entries can not be ~e.member of ~ ulA: · ·r
 = {x e V: Ax = 'OJ
•NulA
~ of T) Range of T = fr(x) : ¥ x e V}
= {Ax :¥xe V}
=Col A ,
:. ~ of ~ear ~.t'onnation Tis NuI A , . .
wl>ere A JS matrix llssociate With linear....__", and_ 1"nge of linear transformation T is Col A
.... Cll.15I<>nnation·r. '
c ...ainple 7: Let T: R3 ➔ R3 be a linear transfonn ti d . -
~ (i) KerT (ii) lmT
a on efuted by T( )_( . d
x, ~, z - x, y, - 2y). Fm
5oJution:
Since,

[A OJ =[~ f g g1
. 0 -2 0 ~J
1 ' 0 () OJ . _.,. _,
-: [:o · 1 _o ·o 1
• •
.; , .
•· • • ,·
o 'o o-o
is reduced echelon form x1 and x2 are basic and XJ is fr~, ~hich are obtained by
x1 ·=0
' 0 ..
X2_= ,..
0 c'
X3 = free
, .· •' , ...
Thus,
NttlA=ff!JXJ~'~}, I - I

And, Ker_T = ffl]= X3 e R}, , ·! Z}f . .. ~

lL~.
J

I };, .' l ., , • ' I ,',

ForColA:

1'ht1s,
 JIATIDMATICS D
. · [ p(O) ] then sholY that T is linear transfO?tna . 1·
Example 8: Let T: P2 ➔ R2 defined by T(p) = p(l) fT tiot\
thekemelo •
and find a polynomial p in P2 that spans

Solution: . . and c be any scalar then

For linear, let p and q be arbitrary polynouual m P2
0
(p + q) (O)] -[p(O) + q(O)] = [p(O)] + [ q~lo = T(p) + T(q)
T(p + q) =[ (p + q) (1) - p(l) ,+ q(l) p(l) q

and

T(cp) =
This means T is linear.
n:~ ~~o =c [~~~o = cT(p)

For kernel T, we have

T(p) = 0 ⇒ [~~~n = 0
This gives, p(O) =0, p(l) =0 .
Therefore, any quadratic polyn~mial p for ~hich p(O) = 0 and p(l) = 0 will be in k,emel of T.

r .__/__HX_En_c_,_s_E_s_.2___________-i_____.l
1. Determine if w
-4
[1] [1]
3 , u = -1 are in Nul A, where A= ~ ~2 .0 .
=
1 . -:-8 4 1
[3 r-5 -3]
2. Find an explicit description of Np.l A, l?Y listjng vectors that span ~e-null space.

(i) A=[~ 0 -2
1 3
4l
-2J
1 -3 0 07
(ii) A ~ [ D O 1 OJ
' [1
(iii) A= 0 .
-4 0 2 · OJ
0 1 -5 0
0 0 0 0 1
3. Find A such that the given set is col A

(i)[,;:
l
:],
2r- s- t
r, s, t, e R}
c+d
(ii) ffb!:~~~d]• b,
ll
,
c~de Rj ·-

4. For the given matrices, find k such that NulA ~d ColA are a subspace of Rk.

2
-1 -6J
3 i '= -27
(ii) .,
-20 -5OJ ~-
(i) A= -4
[
3 -9
12 0 [-5 -5 · 7
7 -2.
' [4 5 -2 6 . 07 1
I ' t

(iii) A = 1 1 0 1 OJ (iv) A = [1 -3 9 0 -5]

5. LetA= [ _
1
-2
~ and w = [;]. Determine if w is in ColA. ls w is NulA? ·
10
-8 -2 -~ 2
6. l.etA=
[
r 2
-1
1
2
6
0
-2

-
o and w =
2
o Determine if w is in ColA. Is win NulA?
 filld a nonzero vector in NulA Vector Spacee W r = - - -
7.

9 -6 (ii) A = -1

8. Let a1, a2, ... , as denote the columns of th _05

5 1 2 2 O] e matrix A, where
3 3 2 -1 - 12
A== 8 4 4 -5 12 , B = [a1 a2 cl.4]
[
2 1 1 0 . -2
a. Explain why c13 and as are in (:olumn space of B
b. Find a set of vectors that spans N ulA. ·
c. Let T : R5 ➔ R 4 be d efined by T(x) -Ax,DoesTo
_ tO
9, If T : R3 ➔ R 2 defined by ne one or onto?

T (xi, x2, X3) = (x1 + x2, x2 + X3) be linear trans1orm.ation

£ • find k rT
10. Let H be the set of all vectors in R4 h · . ' e and ImT.
d - w ose coordinate a b c d .
d an c - a - b. Show that His a subspace of R4 • ' ' ' satisfy the equations a_ 2b + 5c =

1. Yes 2. (i)

(iii)
4
1
0
-2
0
5 3. (i)
• I
. (ii)
[1i_-1~ 03 1
-3
1 0 . 1' 1
0
0 0
(ii) k = 3 for NulA and ColA
4. (i) For NulA, k =2, for ColA, k =4 I· . •

(iv) k =5 for ~ulA and k =1 for ~olA

(iii) k = 5 for NulA and k = 2 for ColA
'I
6. w is in NulAand ColA . ·
S. w is in ColA and NulA

~ ~umn
, p • I

(ii) [ i] is in NulA, of..A is~ nonzero vector.in ColA.

-1 . ~ . . \

-1/3 -10/3
-1/3 26/3 (c) Tis neither on~ to o~e nor onto.
1 0
S. (a) [B, i13], [B, as] is consistent (b) 4
0
1
0
 IIATBEIIATICS-0 .

5.3 LINEARLY INDEPENDENT SETS. BASES

. Proof: Suppose that the vectors ·{v1, v2, ... , vp} are linearly dependent then there exists a set of scalars
Ct, C2, ••• , Cp of which of least one say Cp ;t Othen CtVt + C2V2 + ... + CpVp = 0.
Since Cp ;t 0, so cp-1 exists. Then,
vp ~ (-ietcp-1)"-vf+-·(:-cicp-1) v2 + ··: ... + (-cp-t cp-1) Vp-1 ·
r:: dtVt + d2v2 + ... + dp-tVp-t where di= Cicp-1.
which shows that Vp can be written as a linear _combination of Vt, v2, ... , Vp,t•
This completes the first part.
I
Again, let Vp can be expressed as a linear combination of Vt, v2, ..., Vp-1 th~ we need to show
that the set {v1, v2, ... , vp} is linearly dependent.
Let,
Exai
Vp = C1V1 + C2V2 + ... + Cp-tVp-t•
Sob
or, C1V1 + C2V2 + ... + Cp-t Vp-t - Vp = 0.
or, C1V1 + C2V2 + ... + Cp-t Vp-t + (-l)vp = 0.
Showing that Cp = -1 ;t Osuch that
C1V1 + C2V2 + ... + Cp-t Vp-t + CpVp = 0. '
Thus, the set {vi, v2, ... , vp} is linearly dependent.

Example 1: The set f(1, 0, 0), (0, 1, 0), (0, 0, 1)1 is linearly independent whereas the set {(1 2, 3),
(2, 4, 5), (3, 6, 8)J is linearly dependent. '

Example 2: Let P1(t) = 1, P2(t) = t, P3(t) = 4 - t then the set (Pi, P2, p 3} is linearly dependent in P
because P3 = 4P1 - P2.
n J . ,
~ :~ ~b$pace of a ~ecm.spaoo '(.-.~--~;!;':l'~ set vecton, B
(i~ :e set {bi, bl, ••., bpJ is linearly hlde~de~t,··•. ·. t{

) Span fbi, bi, ..., bp}.

..
.,
 le 3: In R.2 the vector V1 • (1 0) Vector ap...
~Pa, the vectors v1 = (1, 0 0) v' 'v2 == (O, 1) fonn ab • a,-§#r-81
Jtl"' ' , 2 == (O 1 0 as1sfora2
staJI cfatd basis.
. ' , ), v 3 == (o, o, 1) fonn ab • .
~
Pl"' fhe set {1, t, t2, ..., tn} is called standard basis f cisIS. These bases are called the

--a1pl• 4! froVe that the set of Yedo or IP'•· ' I

~ rs (3, 0, -1) (0
sa"'ti•"' ' . '. , 1, 2), (1, -1, 1) form
d_,. we have to show that v v a basis of R'.
p~- . - 1, 2, V3 are linearl · d '
for u.nearly independent, Ax -= 0 · Y m ependent and th
~ ~ -1i ~] "e~:p~R•.
[ -1 2 1 Q . V (

~[ ~ ~ !1 ~l~J ··
0 0 -8 •r lr . ~

No basic variable so having a trivial solution.

. Thus
' ' · -V1 ·v . .
rs for Vi, V7J V3 span R3 ~ , 2, V3 are linearly in<f'1pe~dent.

•
A=[~-1 . 2~ -1i]
1 ..
.-
·1
l ,. I

~[~ ~ _;J
0 0 -8 .
F.ach row has pivot so, column of A span
respectively. So, {vi, v2, v,} span R' . . ., R3
· That · v,, v, and v,
15
1;""• pivot 1, 1 and -8
w
Thus, {vi, v2, V3} is basis for R3. l I
, ..J .

Example 5: _ Let S = (1, t, t 2, •••, tn). Verify ~t Si' a basis for II'~. ·
Solution: ". .. ' -
Oearly S spans IPn• Now, we have to.shQW .that S is linearly independent for this. Let
Co. 1 +Ct. t + .... + Cn. tn = 0 (t)_. ) . .. .
This sI:iows that the polynomial on the left has the same valu!'5 as the zero polynomial on the
right and we know that zero polynomial in II'n means more than n zeros. So, it holds ii

Co =Ct = .... =Cn =0 for all t.

Therefore, S is linearly independent. Hence, it is basis for IPn•
 pTDJIATICS-U
6
OJ [2] • [ 16] and H. span (vi, v2, V3}. Note that v3 == 5"1 ..,
Example 6: Let Vt • 2 , V2 = 2 f' V3 -5 , ~\
[
-1 0 _ (v vzJ. Then find a basis for the subspace b
and show that Span (vi, v2, v3J - Span 11 •1.

Solution: for any v e span {vi, v2}

Every vector in Span {v1, v2) belongs to H because { }
+ 0v e span v1, v21 V3
V =C1V1 + C2V2 =C1V1 + C2V2 3 } .•.... . . .. ... (i)
⇒ span {vi, v2) c span {v1, v2, v: v Sine~ V3 = Sv1 + 3v21 we may substitu
Now let x be any vector in H-say, x = c1v1 + C2V2 C3 3· _ te
X = C1V1 + C2V2 + C3(5v1 + 3v2)
= (c1 + &J)v1 + (c2 + 3cJ)v2
= d1v1 + d2v2 e span {v1, v2).
. ... .... (ii)
Thus, span {vi, v2} c span {v1, v2}
From (i) and (ii)
span {v1, v2} = span {v1, v2} ·
It follows that {vi, v 2} is a basis of H since {v1, v2} is obviously linearly independent.
Bases for NulA and ColA

r
t
- To find the bases for Nul A, first we find the spanning set of Nul A and then check this
spanning set linearly independent or not. If linearly independent then it is basis for Nul A.

Example 7: Find the basis for the null space of the matrix A = ( 1 2 43 ).
2 3
Solution: ., ..., r•
We know that Nu.I A = {x e R : Ax = O} so for solution of Ax = 0
3

1 2 3
Augmented matrix of A = ( ) is
2 3 4
1 2 '3 : o·)'.:~(.1 o -1 \ ,o)
( 0 -1 -2 : 0 0 -1 -2 : 0
st
Pivot is in 1 and 2 column so ~1 · and x2 are basic variables and X3 is free variable. Hence
nd

Ax = 0 has non-trivial solution. For solution . J

st
From 1 row, x1 - X3 = 0 ⇒ x1 = X3
From 2 row,
nd - x2 - 2x3 = 0" ⇒ X'.2 = - 2Xa , :
and X3 is free variable •1 ,• • , ' I .I

Thm,roluti~x= [:J=[-: ]=x{-:2 J.

l
Thm,NWA={[-: :ae ~}mdh=[ ~2 j~PmdspmNWA
Henre, {[-:2)} IB b~ fur NW A.
 Vector 8pacea W ~ - -

2 8 l 2
5 20 2~

.
I 'J [1 4 0
~ 001 -21
8
OJ
0 ,- '
0 0 0 0 .
0 0 0 0 1 ' IS a reduced echelon form

mr
Here, pivot column of A is 1st 3 d O ·

Th~, ~IB for Co~= {u) mr cru: :m

example
. 9: Find a basis for the set of vectors ml3.
. h
Solution: , , , mt e plane x _ 3y + 2z = 0
Given, ' r

c:k this 1,

A.

⇒ AX=O

So, (A 0] ~'"'(1' .:.3 · :l

Here, Y and z are free variables."

So,
~ • r
·
r:
X
[X] [3] ·~ [-2]
"'

= ~ =y ~ + z ~ '
~ I :t!:.J "
f

~[n [!]}IBbams. ..
t• r • u ( .
.., • I

Hence

.l_EXERCISE 5.3 lt
'I r
 IIATBBIIATIC&-11

3. Let Vt
[4
1
= -3 ,
] V2 -
6
_ [ 2]
-1
, V3
= [:i] =[1] .
3
, V4 9 Find a basis for the subspace W soan.....11\:Q. ½
a - .....

. ( ;

Then . every vector in H is a linear

4.

combination of Vt and v2 because

mm =s + .[[]. ffi lvu v,} a hams for H?

5. Find the bases for the null spaces of the matrix

[~ ~ =! ~]
3 -2 1 -2 i
6. Find the bases for Nul A and Col A · .. ·
2 3 -4
-2 4 -2 -4] 2 0 2
'

a. A=
[
2
-3
-6
8
-3
2
1
-3 b. A=U 4
6
-3
0
10
6

I
7. Find a basis for the set of vectors in R2 on the line y = -3.x.
8. U,t v, +H =UJ =[f] v, v,

Sv2 - 3v3 = 0. Use this information to find a basis

and also let =
H span Iv,, v~ v,} it

for H .
~ be verified that 4v1 •

111] . . .
1. (i) Since the given matrix A = [ 0 1 1 has 3 pivot positions. So by the invertible matrix theorem, A is
0 0 1
3
· .
. · · . .
invertible and it's columns form a basis for R .
(ii) It is not a basis for R 3. (iii) It is not a basis for R3.
(iv) It is not basis for R 3. (v) No (vi) Yes
(vii) No (viii)
. No.
2. {v1, v2} is not basis·for R3 and R2. 3. {vt,'V:i} 4. Yes

5. &filsforn~A-mJB]}
6. (a)~~RX.~l: =HJ [t]}. eu~fornull~,~[ ~1 2].[-~ 2]} ,

-2 . -2
1 0
0 2
0 1
,o 0

8.
The set {Vt, V2}, {Vt, V3} and {v2, V3} are linearly independent and thus each forms ~ basis for H.
 Vector Spaces W l§w:# 51

$-'~~===~=~~======~==~====~~
lr I

foOJIDINATE SYSTEM

P--rt.·se B {bu b2, ... , bn} . a basis

== . for v and
is
5i1pl'° din3te of x) are the weights c
x• . V
ism · The cooiclinates of x relative to the basis B (or
.i.e s,coor
P'
1, c2, ... , Cn such that x =• cibt + c2L12
1-._ +
• •,
+ CnVn•
1..

Ct, cz, ... , Cn are the B-coordinates of x, then the vector m

. Rn
If

• 1

[x]-e =

Cn

is the coordinate vector of x (relative to B) or the B coordinate vector of x.

The mapping x ➔ [x]s is the coordinate mapping (determined by B).

f.XaJIIPle 1: Consider a basis B = [b,, b,] for R2 where, b, = [!1 :U.d =ill suppose
b, x in R' has

[-2]
the coordinate vector [x]u·=· 3 . Find x.
\:.. . 1
.
Solution: .u

Here, • l l •

~
·
-- ~ ~

~--~
. -
I!] are the co~~~~ of.x relative io ~e st
• .l •

n, Ais Example 2: The entries in the v ~ ,c . =

basis £ = {~ e2l . •~ • ,J • • ·-

' .
Solution:
Since

then there exists scalars t1, ·· .,

B span V,
V So
t r space • ,
Proof: Let B = {bi, ... , bn} be a basis for a vec o ,• .
j
Cnsuch that ······· (1)
 holds, for x in V. If possible, suppose x also has the representation

.r = d1bi + ... + dnbn ········ (2)

For scalars d i, d2, ..., dn. Then subtracting (2) from (1) then we have,

0 = (Ct - d1) bi+ ... + (en - dn) bn ·· ···· (3)

Since B forms a basis for V, so B is linearly independent. Then the weights in (3) must all ~
zero. That is, 9 = d; for 1 ~ j ~ n.
501
This means the scalars c1, .. ., Cn are unique.

Example 3: Let bi = [i], b 2 = [-11], x =[:]and B = {bi, b2J. Find the coordinate vector [x] 8 of x
relative to B.
Solution:
The B<oordinates c1, c2 of x satisfy

c{i] + c{~1] =[:]

⇒ [i ~1][~~ = [:] ......... (i)

This equation can be solved by row operations on an augmented matrix or by using the inverse
of the matrix on the left. In any case, the solution is c1 = 3, c2 = 2.
Thus x =3bi + 2hi and

. . .
- The matrix in (i) changes the B<oordinates of a vector x into the standard coordinates for x.
An analogue changes of coordinates can be carried out in Rn for a basis B = {bi, "2, ... , bn}.
Let,
Then the vector equation is
X = c1bt + C2~ + ... + Cnbn
is equivalent to x = Pe[x]s.
We call Pe the change of coordinate mabix from B to the standard basis in R".
Pe[x]e =x and Pe-1x =[x]e.

Example 4: Find the change of coordinate matrix from B = [~. [·;] to standard basis of J.'.
Solution:

Let bi = [D = c,ei + ciei where ei = [~], fl =[~] , E ={ei, ei} is a standard basis for R2•
⇒ Ct= 1, C2 = 2
 \
Veator Bpacee W IC&Ama51

'I
1 I •1
,
The set B = {1 + t, 1 + t2, t + t2} is a basis for P 2• Find the coordinate vector of
p(t) = 6 + 3t - t2 relative to B i.e. find (P]8 •

sotution:
since P(t) e P 2 and B is basis-so let coordinate of p(t) relative to B, say c 1, c2 and c 3.
Then
c1(l + t) + c2(l + t2) + c:3(t + t2) = '.p(t) = 6 + 3t- t2
,., ,.. => · •c1 1+ 'cit+ c2' ~ c2t2 + c3t + C3t2•= 6 + ·3 t- t2
=> (c1 + c2) + (c1 + C3)t + (c2 + c3)t2 = 6 + 3t - t2

. .... (i)

verse
ri,. J I

Thus, the au gmented matrix of system!(i) is

J .

u 1]
1 0
0 1
1 1

>r x.
( 1 -- •.
l~--;c ~]
1 0
0
1
1
...,
- f f : '.

rn
1
1
-1
0
1 !]
1 . -3
t .

[~
0 -1
1 1
0 ·2
!] , . ,nu• . :''

0
1
-1
1 J-1 7} . ,
?(,.fi
' \

rn
,J

0 1· -2
,.
, . '

. , I
')

rn
0
1
0
0
1 t]
2
⇒ C1 -5
- , c2=1and c3 == -
 MATBElfATICS-ll .

Hence, [Pl• ~ [~] {!]

. t erify that the polynomials 1 + 2t3, 2 + t - 3t2, -t + 2t1
Example 6: Use coordinates vectors o v , . .. tl
are linearly independence.
Solution: · ., · .
.
The coordinate vectors are (l , O, O, 2), (2, 1' .:.3' O) and (0, -1, 2, -1) respectively. Let A be a 11\ai-..:.
... U\
using them in columns. Then
1
0 -1
2
1 0] [1 ,0 0]
0 1 0
A= 0 -3 2 ~ 0 0 1
[
2 , 0 -1 . 0 0 0 .
which shows that A has pivot in each. columns. So, the given polynomials are linearly
independent.
J 1

Example 7: Let v1 = [~, v, =[ 1], x = [❖] and B '. = (vi, v,J. Thell B is a basis fo,

H = Span {vi, v2). Determine if ; is in H and if it is find the coordinate vector of x

relative to B.
Solution:

H x is in H, then the following vector equation is consistent. · r

I r ,

The scalars Ct and c2 if they exist, are the B-coordinate of i. Using row operations we obtain
-1
0 3] [1 0 2]
12 ~ 0 1 3
'

1 7 0 0 0

This implies, Ct= 2 and c2 = 3 and [x]a = [;].

Thus, the coordinate system of x on H determined by B is (x] 8 = [ ; ] . .

Coordinate Mapping
Let V be a vector space over the field K and B be a basis for V. Since for each element v e v there
exists unique coordinate vector [v]a w. r. t. the basis B. Since the representation is unique so there
is one to one correspondence between the vector space V and space subspace of Rn which is
called coordinate mapping.
. ___
[ ·]s __,,
X

V
 ... 1C-aAmit st
Vector Spaces m r : : - - - -

l~Xl~llCISI~ 5.4

. find the vector x determined by the given

. coordinate vector {x}s and bas' B
1

(i) s=[!s} [i].1x)e=m (ti) B={lH l~J m[;J[x]e=m

(ili)B=[H mHl[x]e=[!]
rly ' I

Find the coordinate vector [x]s of x relative to the given

. basis B = lbi, bi, .... bn}.
2.

or (i) bi = [ ~1 bi= [!;J X =[ =!]

'

3.
.
Find the change of coordinates matrix from B to the s~dar: db'as1s· ·
.
L .-· ID Rn

4. Use an inverse matrix to find [x}s for the given x and B.

5. Use coordinate vectors to verify that the polynomials 1 + 212, 4 + I + St' and 3 + 21 are linearly

2
dependent in P2.
6. Let B = {l + t', t + t', 1 + 21 + t'I is a basis £of P,. Find the coordinate vector of p(t) = 1 + 41 + 71

2
relative to B.
The set B = {1 _ 12, _ t', 1 _ t + t'I is a basis for P,. Find the coordinate vector of p(t) = 2 + 3t - 6t
7. 1
relative to B.
2
8. Let pi(t) = 1 + t2, }'2(t) = t - 3t2, p3(t) = 1 + t - 3t
1 mials form a basis for P2.
(a) Use coordinate vectors to show that theseF'Pod yno
q in p2 given that
f P
(b) Consider the basis B = {pt, p2, P3} or 2· tn
 MA1'BEMATICS-D

a. Show that the set B ={bi,~ bJ} is a basis of R3•

b. Find .t he change-of-coordinates matrix form B to the standard basis.

c. Write the equation that the relates x in R3 to [x]e.

d. Find [x]e, for the x give above.

1. (i) [!7] (ij)

[}1 (iil) [~]

2. (i) {~] (ii)

~=n - (iil) [-i]
3. (i) [_1 -~J (ii)
[3 2 -}]
0 2 4. (i) [i] (ii) [ ~]
6 -4

6.
GJ 7.
[ll 8. (b) q(t) = 1 + 3~ - 10 t2

9. [x}e =[fl
f -Y, .!

□□□
•. {
I,-

G ~- • r t' .~ , t ,. - i

.. )- '
 • I

• I

Cl}RPff.1\

...

.. .. • 1 ~ ,. • • • • ' ..
~!C I '

t ·1 ,;
VECTOR SPACES
CONTINUED ' I'

LEARNING 0111c0Jiti's ·,, ~.~-. '~- ~-·~

. . ~~~- ~~~
t' ! J l

. · . . ii should be:able to: ') .....

After stadylng'tlds cupter JO - • I I
, Space
a. The Dimension of'a Vedor
a. Rank
a. Change of Basis
• e Equations
a. Applications to Differenc
-'l:!l. Application~ of Markov. chains
 JIATBBIIATICS-U

IIWRODIJCDON
. ension of a vector space, rank of a matrix, change of
This section introduces the concept of dim f ector space Also the section highlights 80,...
bas·15 to another basis o a v · ~,e
coordinate matrix from one • that are best explained using linear algebra.
fundamental properties of linear difference equation

8.1 ftE . JIIMENSION OF A VECl'OR SPACE

. .
•· i :. } is a basis of v then the number of basis
· Let V be a vector space over the field K, let {v1, V2, • • ., Vn • •
elem~ ts is ~enoted.by ~ V and called dimension of the vector space V. In this case, we write
dimV=n.

Example t: Since the set [~, [!} [:] in R2 is linearly dependent because the set containing 3
v ~ but the dimension~of R2 is~- :- . ..

Remar.lc: 1. In a vector space, there may be different bases but their -number of basis elements must
be same and that number is called the dimension of vector space.
2. Ha non-zero vector space Vis spanned by a finite set S, then a subset ofS is a basis for V,
by the spanning set theorem.

The standard basis for Rn contains n vectors so dim Rn = n. The standard ol mia1
basis {1, t, t Jsuch that dim P2 =3. In general dim i.>n =n + 1.
2
p YI?~ ~
. Example 3: The space P of all polynomials is infinite dimensional vector space. I

i
~pie 4: Let H • S pan {ui, u2}, where llJ Vectm 2}l'-.-ollltl•Lla1111w.,.d. W _.-.,..:.;~·
~- C
oet·l8 lineady
• Independent. u,= 2
So, dim -
3 'Hu,=•2. o3 Then \u1, u,\ Is basis I!"~ beca- the

.
. d ofColA 18
'
m the dimens1ons
A= 1
.
-3 6
of th
1 the number of pivot
enull space and th
-1 1 -7
-2 2 3 -1
e column space of

[ . .
: 2 -458-4 .

Solution:
Row reduce the augmented
· . matrix . .. .
· [A O}. to echelon form·
~ [1 -2 2 3 -1 0 .. .
0 0 . 1 2 -2 0~ I

There are thr o o o o o ·o . l : ,,,. '

• ee free variables x · •
ColA is 2 because A !las
· . two pivot x, and >:•· Hence
•" 1:9\umns.~ . . the ,.,~, . of. NulA is 3. Also, dim
_,.erunon

£xantple 6:
- . of the subspace
Find a basis and dimen&lOft
.
- .
• . H = \ -9a
,
1 1
6a-
3a ++lb
Sb--+2.c
6b
-3a+b+-c -
c3c a, b, c e R ~ ~
,-. r -
- •r.• • • •• • I

Solution: -~·
We have,

[ 6a-2b-2c
~+6b-c
-9a+Sb+~
1= a
r~ r1 r~
6" -
~~ +b
:;;;2
~
. .:.2
+c ~ -

-3a + b + c' I

av1'+ bv{ + cv~ \

I •
= '
,.
-1 l •
I ·,
6
3 -2 . \ IL,.

6 -2 V3:::
, V2 = 5
3 " • .. I
where, v1 = -9 1
1
which shows-3that H is lineal combination of vu v2, v3. O~ly, v1 * 0, v, is not a multiple of vu
but V3 is a multiple of v . So, by sp~g set theorem \vu v,l also sp~ H and sinee it is jjneatl I ;

1
independent So, it is a baSis for Hand dim H"' 2. •. · ·
 MATBBIIATICS-U

_1 EXERCISE 6.1
1. Find the basis and dimension for each subspace }

~
2a 2c ]
a-b ceR
(i) Jr-4b] a, be R} (ii) b _ 3c , a, b,
lL-2a a+ 2b

(iv)
{
-% :~
P - 2q ]
p, q, r e R.
}
(v)
[s -t]
s; t s, t e R
}
-3p + 6r
(vi) {(a, b, c): a - 3b + c =0, b - 2c =0, 2b - c = 0} '
2. Find the dimension of the subspace of all vectors in R3 whose first and third entries are equal.
3. Find the dimension of the subspace spanned by the given vectors

t (ii) [ Ht}fi}[1]
~
·II 4.

(i)
Find the dimension of NulA and ColA for the matrices

A=[~ -: ~ ! '~J
0
0 0
0
0
0
5
0
2

1
0
1
.: ·
(n) A
t ['t
O9
= O ff" 1 · · ~
~ 5·]
•

(w. .)A-[3
•

2]
·- ·~ 5
I

(. ) =r;2 4 =2 ~:-41
1v A
-3
-6
8
3 1 _
2 '-3
;r~
(v) A _
· r~r. ~ -=~!~ !~-;J ,;.
- 1 2 0 4 5
3 6 . -5 19 -2

1. (i)Ba=[U ftJJ Inm=2

2.
· (v)

dim=2
ffiI fiJdun= 2
(vi) Nul A ={O}, and dim = 0
I , 3 (i) dim= 3
4 (') • - . . . .
• 1 dim ColA = 3 dim NulA
, _
-2· (ii) dim= 2
(~) dim C~IA =2, dim NulA =2
• • , I

(iii) dim ColA = 1, cfim NulA = 1

(v) dim Col A = 3, Nul A ;;: 2 (1v) dim Col A = 2 dim Nul
A:-
Ull!l ' A=2
 Vector flp CN Continued W lCeAma 61

'
Find the bases for the
A [;l -:
= ! •rce~ ;]e column space, and the null space of the 11111trix:
3 11 -19 7 1
1 7 3 5
Solution: To find the b c th-l -3 ·
' ech c · ases i.oi: e row space and the,column.snace. We have to reduce A to an
e1oni.orm.. ·· ., F ,

1 3 -5 1 5]
0 1 -2 2! -7·
I •

A~B=o O O -4 20·
[
. . Cl Q . 0 . 0 . .0
Therefore, basis for row space-of the matrix A
is {(l, 3, -5, 1,. 5)~(0, 1, -2, 2, -7), (0, 0, 0, -4, 20)}. .
For the column space observe 'f rorri 13 that the pivots are in columns 1 st, 2nd and 4th • Hence,
columns 1st, 2ruLand 4th of A (nof B) form a basis for Col A. ·
Therefore, basis for Col A is

Therefore,
X4
XS

-1 -1
2 -3
1 0 is basis for·NulA.
Thus,
O 5
0 1 I '

an Nul A and are linearly independent)

(Because these two vectors sp
 fi$''
column 1,-
,.. ~

RanJcTheottll1
The djmenSion of the column space and the row space of an _m><n ~~~are eq'Ual.
common dimension, the rank of A, also ~uals the number of pivot positions m A and ~ -~ lf A
the equation . , ~- StJ
=n . 4. ~
rank A + dim (Nul A) . .
&ample 2: (a) If A is a 8 x 11 matrix with a three-dimensional null space, what is the rank of>.?
(b) Could a 5x8 mabix have a two dimensional null space?
Solution: · (i)
(a) Since, A has 11 columns, so
Rank A + 3 = 11.
⇒ RankA=S
(b) No, if A is 5 x 8 order matrix, then dim NulA can not be 2, if it is then, Rank A should be 6
which is impossible since each columns are from R5•
(ii
- dim Col (A) = dim Row (A) =Rank (A) =Rank (At)
fm,ertfble Matrix Theorem '{Continu
ls A be nxn mamx. Then• ·
A is an inverbole ·
(a) Thecol
{c) DimColA
(e) NuIA'. • {O}
(

~ EXERCISE 6.2
1. Assume that the matrix A is row equivalent to B without calculations. List rank A and
dim Nul A. Also find bases for Col A, Row A and Nul A.

-4 9 0 -1
(i) A=G -6
2 -4
10
fl,B=n -2
0
5
0 t]
6 -6 6 3 6 . 2 6 -6 6 3
(ii) A=[~-2
-3
9 -12
6 -3 0
9 3
-6
12 IB=
0 3 0 3 3 0
0 0 0 0 3 0
3 6 3 3 -6 0 0 0 0 0 0
1 1 -2 0 1 ' -2 1 1 -2 0 1 -2
1 2 -3 0 -2 -3 0 1 -1
(iii) A= 1
0 -3 -1
-1 0 0 1 6 ,B= 0 0 1
1
1 -13 -1
-2 2 1 -3 0
1 0 0 0 0 1 -1 l.
-2 1 0 2 - 0 0 0 0 0 1 4.
 2.
Find rank A dim Vec:to.r 911 r o.. Oontbia4,W §-4
, NuJA and bases
2 -1 for ColA, Row A and NulA from the given matrix
1
-~
~ . .A= 1. -2 -4 -: 82] .
fi~ [ -7 8 1 -
0 3 -10 .
4 -5 -7 • 0 • I
3. H 4><7 ~trix A has 3 ~ . .
4. Suppose a 4 x· ~- Fmd dim NulA, dim Row A, and rank AT. •
7 matrix, A has four pivot columns, Is ColA ="'R.4? Is NulA = R3? Explain' yo ur
answer.

I I

1. (i)
RankA = 2, dimNulA • 2;
1
f , 'I
[ !1],
s•
[i] -6 >
i~ .basis for ColA;
rs l

[t} [TI is
I I I • F

1(1, 0, -1, 5;, (0, -2. 5, -6)} is basis for Row A. basis for NulA

(ii; ~A ~~-~ •~L=; ' . -

Basis for row A= {(2; 6, ~, ,' 3; 6)1 (0, 3, o,·3, 3,·o); (0,
Basisforcoi'A--111
·o, o, O, 3, Oj} 1
'
[t1 [il
• • '

0 i O. -3 ' Iv · • r
0 -1 0
1 0 0
Basis for NulA = 0 , 1 , 0
0 0 0
0 0 1
., I
(ill) Rank A = 5, dim NulA = 1

·-m~~={[HtU[fJ[!l[!]}.
Basis for row A= {(1, 1,-2, 0, 1,-2), (0, 1,-1, 0,-3,-1), (0, o/"°i, 1,-13, '.;1), (0, 0, o,"o, 1, -1), (0, 0, 0, 0, 0, 1)}
-1 • I •
j .r •
-1
1 '.
-1
Basis for NulA = 1
0
0

2. Rank A = 2, dim NulA = 3; -fo•~i=i}1 l!]

-2 5 -6 . . l
-3 4 -4
)} 1-. 0 0 is basis for NulA.
Basis for row A= {(1, -2, -4, 3, -2), (0, 3, 9, -12, 12 ;, 0 1 0
0 0 1
• I
r.
3. Row A=3 RankAr= 3 7
•
DimNulA -= 4,dim ' dim Nul A~ 3 but NulA is a subspace of R •
4. Yes, No, NulA ~ R3, it is true that '
 -•~GE OF JIA§IS
,,_:,€~ - ~
~
fffe
If there is two: } :
tlt
1 o,dered bases fur an n- dimensional vector space V, say
v e v . How are [v]e and (v ]c related ? The tool we need in
C = {ct, c-i,_· · "change of basis vetor. .
B lb,=
~
to da '~
~t··, ~,
efficiently•~ e bn} and c = {c1, c2, ...., en} are basis for Rn. Then change of coor dinate . \
Let B = {bt, Vu • •••, p II\at:rix ~
to c is denoted by c:B and defined by Ct-B = [[bi]c [\,l]c .... [bn]c] i
and, p
[x]c = Ct-B [x]e
p
It means, the matrix Ct-B convert B-coordinates into C-ordinate.

Note
p - [ p
· Ct-B - Bt-C
1-1·

Example 1: Let B ={~ h?I and C = {ci, c2I where bi = [-;J, b2 = [:!], c1 = [ ~], c2 = [ !s1 are \ht
two basis for R 2, then
(i) Find the change of coordinate matrix £,om B to C.
(ii) Find the change of c4:>ordinate matrix from C to B . ·
Solution:
(i) For the change of coordinate matrix from B to C
P ·
[[bile Cbllc] = c~B
For [bi]e, Let bi = X Ct + y C2

⇒ [ -9J
1 = X [ -4
1] -+: y-[·-5'
3. J 'I 1
I

i.e. -9 = X + 3y . ........ (i,

and 1 = -4x - Sy ........ (ii) . .. - ,,
Solving, we have, x = 6 and y = -5:
~erefore, fbt]c =[_~] ·
Again, for fbilc , let ~ = x c1 + y c2 • I

⇒ [~n =x[!]+y[!]
i.e. X + 3y = -5 ...... (m)
-4x-5y = -1 ....... (iv)
Sol~g, we have, x =4 and Y =-3.

Therefore, [l>z]c = [ ~].

Thus,
p
C~B =[[bi]c [bl]c] = -5
[6
(ii) And,
-4]-[-3/2
6 - 5{ 2
-21
3 .
 Vector8paceaca, •ta.w.4

~pie 2: Let bi • [ 1,._ =

and C = (c,, <2}vz
then -31]' [-2]
4 • = [-;J
9 • Ct [-SJ
<, = 7 are the bueo of IV given by B • p,,, t,,\.

(a) Find.the
. change of coordinate matrix fro C
(b) . .Find the rlumge of . • , m lo B.
Solution: coordinate matrix from B to C.

(a) . y elemen~ row operations

We have to appl
(bi 1'2 : Ct c2] ➔ [lo O :
1 :
X
y
x'y' ]
e Then, p
B+-C= y
[x x'y' ] ·
Here, [b, _!>, _c! c,] =[:3 ::{ '. -; ~ -[~ ~ : !]
Therefore, B+,C = 6
p -. rs ' 3] 4 .
I

.
~
Using first part, .
f ,t! •
i ,• pl·-:=['·B~~
G~ B
.~1ri-N•~![~ '-3i\ -~[2 --•'-312lll
. -~ 2 ~ 1<~15.,j -3 . '5/2.itl ·, I.

~p!i,.3:·. P>A5ilk.! ,:wo base•. B, = .(b,, !>,I, ~d <;. =c (c,, c,}. for ~ v ~ ,pace V, such tl!at·
b, :,. 4c + c, and b, = - 6c1 + c,. Suppose x = 31>, + b2 i.e. {x}a .-'( f} Find \x}c- · ·' '
1 I • .. t -, • • • • .( I .J ~ I

... I

Solution: p •••••• ( }~ J 1 '. ~ , I

[~Jc.~ G+-BJx}e.., . ;-_ ,
We know that
• •✓- p • I

C+-B = [[bile [bl]d

and
Since,
b, = 4c1 + c, i.e. (b,]c = [!]
i,, =_6c, + c, i.e. [b,]c =[~]
and .

p r4 -6] . all d
1 15c . e
_i.. ....
u~•
nge·of coor~te matrix from B to C.
..
. •
. ·
.. ,!' .·
·
Th~;c~s=L1
From (1)
 JifATJIBIIATics-11
TO DIFFERENCE EQIJATIONS
6.4 APPLICATIONS ~ 0''
.
s«erence equati0n5 are
often the appropriate tool to analyze' discrete or digital data.
• al 1 • •
~
-...l \ti\.,_
,
0

Dw1 . . used to model a continuous process,.a numenc so ution IS .often -~\

dif{erentiaJ equation is . Ptoc1\l1:
,n a related difference equation. l!I\
:S section highlights some fundamental properties of. linear diffe~ce equation that are
explained usii'lg linear algebra. ~
Discrete time signals .,
Let s be the space of all doubly infinite sequence~ of numbers {yk} = {... Y-u Y-1, Yo, y1, y
A signal in S is a function defined only on the infegers· and ·ts visualized as a sequence of nuin~
say {yk}, '

Figure shows three typical signals whose general terms are (-7)k, 1k and (-1Jk-, respectively.

yk= 1k Yk = (-l)k
I

-2-10
IlUilI I -2 0.

Digital signals obviously arise in1 ~l~trical and · conq_ol- ,system ep.gineering, but discrete-data
sequences are also generated in biology,,physic~,-~onomics, demogi:aphy ·and ~ y other areas.
Linear independence in the space s of signals
we'consider' a sefof 01:111 three ·signals iii s; say; '{tik}/ {vk}~and {w1i}! Tney·are nhearly-incieperident
precisely whe11 !1}~ eq~atj~n ;.. :- .- ~ : - . ,_. 1 • nr •. ; .
Ct Uk + c2vk + C3Wk = 0 for all k .... (i)
• J

implies that Ct= C2 = C3 = 0

Suppose, Ct, C2, C3 satisfy (i). Then the equation in (i) holds fof any thrbe consecutive values~of k, say
k, k + 1, and k + 2. Thus equation (i) implies that c1 Uk+ 1 + c2 Vk + 1 + C3 Wk+ 1 = 0 for all k.
and
c1 Uk+2 + c2 Vk+2 + c3 Wk+2 = 0 for all k.
Hence, c1, c2, c3 satisfy

[u~:1 v::1
Uk+2 Vk+2
~:J [~:]
Wk+J C3
= [~]
Q
·for all k .... (ii)

The coefficient matrix in this syste~ is called the casorati matrix of the signals, ·and the d e ~
of the matrix is called the casorotian of {uk}, {vk} and {wk},
H the casorati matrix is invertible for at least one value of k, then ~uation (ii) will unply tba1
c1 =c2 =c3 =0, which will prove that the three signals are linearly independent
 Ji%af11ple _4: Verify that 11t (-2)1t Vector a.,__ "-
, .,and3kareline"'-' . ._...~tta§mat4
5oJutiQn: _ _.y Utd ----
, · · . ependent ·
We have, · · Btgnals.
Uk==}k V _(
By casorati matrix we h

f '4(f• t
.-~... Ute
ave

Vtc Wk~
' k - -2)k, wk == Jk

[
1
k

'
taking k = 0, we h~ve,
'-• 1Utc+1
U•- 2
1<;+
lYk-t-1 . W
Vk+2
Wk+
. k+l ::: 1k+l
Jk+2
(-2)k
( 2)k+l
-
(-2)k+2
3~:}] , I
.
Jk+2

.
-. -.. n~·!J ~ [~ ._; ~J -[~ _; ~ J
>. ~• ~
Smce, each columns are .
9
. . , .,, • , . .
0
O 10
3 8 0
-
.
f.,f; I
1.

pivot columns which h . , . .

Thus, 1k, {-2)k and 3k are line I . s ows that fl:le ~asorati matrix is invertible fork = 0
• • ,., . ar y independent.
.. *· .
r. ,'1 l

Solution:
Given eq1:1ation i$,
:- 1 • • ;: • ·-
1

Yk+3 - 2yk+2 - Syk+1 + 6yk :=-0 for all k ✓ • •• • • (1)

Putting Yk = rk in equation (i) we get ... . ,
rk+3 - 2 rk+2 77 5 rk+l + 6rk = 0
. '

⇒ rk (r3 - 2r2 - Sr + 6) = 0
•' I "J\

⇒ rk (r - 1) (r + 2) (r - 3) = 0 .. . (3)
which shows that rk satisfies the c;fifference equation (1) iff r satisfies equation (3) fi:om (3) thus
lk, -2k, and 3k are all solution of equation (1). .,
' • I

. ".
6.S APPLICATION OF MARKOV OIAINS
,r, • .

The·Marxov chains described in this section are .used as mathematical models of a wide variety of
situations ·in i,iology, business, chemistry, engineering, physics and many others. The model is used
to describe ari experiment , or measurement that is -performed many times in the way, where the
outcome of each trial of the experiment will be one of several specified possible outcomes, and where
the ootoome of one trial depends only on the immediately preceding trial. · ·'

.,
.t • r
 cs-II a.sured each yffr, then a Vector
~ 1'1 . sabUfbs were roe ~
~ tiOfl of a city and Jts . . and 40% in suburbs. The d ~ .
le if the papula ulation lives in oty \t\
for exaDlP ' . that 60% of p<>P . f the region.
[o.60]
as"° "" o.40
could indicate
use they account for the en
tire papulatton o
. called a probability vector.
"° add up to 1 t,eca . es that add up to 1 15 • ' trix
with nonnegative entn . called stoeha&tiC ina .
oefinition: A vector are probability vectors 15 ' ther with a stochastic Inatfix 'ti
. whose columns x x2, ... toge .,
A square rnatnX obability vectors Xo, i, ) , • .
. . a sequence of pr . . _
A Markov chain IS -Px X3 = Px2. .... d difference equation X1t+1 - pxi. fot
such that x1= p"°' x2 - i, • . b the 1st or er . -
"''" is descnbed Y
ch~·
Thus the Markov . . ·ven by the ma~ •
k = O, 1, 2, .. •· city.· and suburbs is gt . ·
. gration between , , .
le 6: The annu al JIU . . , •
ExamP From City Suburbs to , • . •
.,. • I I .......

ro.9s
M = Lo.OS
o.03] city
o.97 suburbs • . d 30A of the suburban•popul~tion
. .

. o ulation ntoves to tlie suburbs, an .:n is 600,000 in· the city and
i.e. each year 5% of the aty ~ p 000 population of the regt . - ,.
. S ppose m the year 2 ' 1 ti in 2001? In • 0021
moves to the city. u .. d'stributionrof the popu a on . ,
400,000 in the suburbs. ~at is the I

~-.rn
. . . .. . .. . .

Solution:- .
ec~~r [. 6()();<looJ
changed to [r.e. ,Xo = 1 "' •
We,saw that after one yeali, the populatton _v '400,??0 : . , .- ,. . . ... ,.. i. .

. · [600,0007 = [ssi:0007·
,[·0_95 o.o3]
1.
0.05 0.97 400,000J 418,000J .. . . . . ./
f 1 milli we find that
If we divide both side of this equation by the to~ population o . on, .
0.95 0.03] [0.6] = ro.582] .
2.
' I
[ 0.05 o.97 o.4 Lo.418 - -. -

-[0.582] . the population distribution in 2001 i.e. 58.2% of the region lived in
The vector x1 - 0.418 gives _ . _ _ _ . :.
the city and 41.8% lived in the suburbs. ·

. Sbnilariy, in 2002,
0.95 0.03] [0.582] [0.565]
X2 = Mx1 = [ 0.05 0.97 0.418 = 0.435
which shows that in 2002. 56.5% of the region ~ved in ~e city and ~~.5% ~ved in the suburbs.•
t

Examplei A small rein<i"te village recei~e~ radio bro~dc~sb ·from· two radio statio~, ·a ~ ews
I ' I •

- --
station and a music station of the listen who are tuned .to the news station, _'JOD/o.will
remain listening to the news after the station break that occurs each·half .hours, while
30% will switch the music .station at the station break of the listeners who are turned
to the music .station, 600/o will switch to Ute-news station at the station break, while
40% will remain ~tening to the m~ic. Supppse everyol)e is listening to the n~,.at
8:15 a.m. ·
a. Give the stochastic mabix that describes .how the radio listeners tend to changC
stations at each station break label the rows and columns.
 ) , b
Vecto_. ■1 • C •k I m-fcawr-fil
· Cive the ini
What
c. tial state wedut. .. ,
Solution: ~ of the listeners will be listening to the music station at 9-.25 a.m.
(a) . Let N stand for n
eWs and M stand for music, then the listeners behavior is given by the table
/ N From To
M
0.70
0.60 N
0.30
0.40 M
So the stochastic ~hix is p., [0.70 0.6]
. 0.30 0.4

(b) Since lOO% of the listeners are listening to the news at 8:15, ~e initial state vector is Xo = [~] •
(c) There are two breaks between 8:15 and 9:25, so we calculate x
2•

=p = [0.70 , 0.6] [1]-[0.70] 1

Xt Xo 0.30 0.4 0 - 0.30

Xl 0.6] [Q.701 =[0.67l

= Pxt ;. ~0.70
L0;3:Q· 0.4 0.30J . 0.33J
.- I • 11 •

Thus 33 % of the listeners ~ listening to music at_9:25. ~ . --1, I..

l •

1. Let B = {bi, hi} and C = {c1, ~2} be .lz~i~Jo_r a vec;tor space V and suppose bi = - c1 + 4c2 and
bi = Sci - 3c2. Find the change of. coordinate matrix for vector space and find [x]c for
x =Shi+ 3bi. •
2. Let B = {bi, hi} and C = {c1, c2} be bases for a vector space V, and suppose bi= 6c1 - 2c2 and
bi = 9ct - 4c2~-. - ,· r -

(a) Find the-clJange of coordinate ~tr!Jarom B to C. ,,.

, . ;
I ••

(b) Find fx]c for x r= -3bi + 2bi

3. . Let B = -{~ bi} and C = {c~~ c2}- ~ bases for c?-. vecto~ y, and suppose bi~;; -2c1 + 4c2 and_'
. i •
~=~-~• l

a. Find the change of coordinate matrix from B to c,- , h ' H.

b. Find fxJc for x = 2bt + 3bi .

4. Let A
· J d B --
= {a1, a2, 33 an
{"- h2t 1'3} be b,ses ,for. a vector space V and suppose
&JJJ . ,• • ,
a1.= 4bi - ~
•

a2 = - bi + bi + 1'3, and 33 = hi - 21'3.

dinates matrix from A to B.
a. Find the change of coor
b. Find [x]s for x = 3at + 4a2 + 33
f} be bases for a vector space V, and suppose ft= 2d1 - d2 + ch,
5. Let D = {di, d2, ch} and F = {ft, f2, 3 .

f2 = 3d2 + ch, and fa = -3dt + 2cb- .

. nix from F to D.
a. Find the change of coordinates ma •
b. Find [x]o for x = ft - 2fa + 2fa.
 ,...TJISll"Tlc&-11 f' d the chal\ge of coordinates matrix fro
\.,..,.l ..,d C • {c,, c,\ be i,,ses for R'· ill tn. ll tot

andBa1sO
1.,et frOIJ\
:= {bi, V t C to B, 1
r-2] p1 {;1 bt =C11 c, = c, =\_\)
b, bt m
• . b, -Q• i,,=[:~ , l·;J-sj~:2\2 C, m
d. 1n-(_t21 ln = c, =li1 c,~I~\
f4l fBl _ 2
c. 1,i==L4j\n"'l4ic1- o =L zJ th b siSJl={1-2t+t',3-5t+4t',2t+3t'\
l9J
• inatri1drot11 e a to
,_ P filld the cb>J'S" of coordin"1es . di(late veetor for -1 + 2t.
7. u•
the standard
2' 1,aSis C • ,l,
,
1
• t2\ · 'fhell find the o--coor ;ng difference equation.
. the so\uti(l[\S of th• accot11P""Y
s. verifY tnat t h • ~ are (ii) 3", (-3}"; y..., - 9y, = o

2•, (-4)';
(i)~UJ U-
Y"'' + 2y••'
"U' -BY•= O . ' (-2)", 3'<
(iii) (-2)", lo
_..;.&u ..1.. .. t the foUo··""g are linearly independent signals
9. V
(i) 1•, 2", (-2)' (-1l", 2•, 3".
(ii) , 2 -.
1
10. Find some solution of the dif£eret1C" equation Y1<+' - Y~• t 9Y• = O
11. On any giV<l1 day, a student iS either neaithY or ill- of the students_ w~ are ~ealthY today, 95%
will be healthy tomorrow, of th• students who are nl today, 55% will still be ill tomortOW'·

a. What is the stochaStic roatrlx for thiS Situation? · · "

b. Supp<>S" 20% of the students are ill on M;,nday. What fraction· ;,r ~eUtage ·of the
students are likely to be ill on Tuesday? On Wednesday;? ; . '. • · .
c. iS
If a student healthy. today, what is the probability that'he or ~e will be healthy two days

from now? · - -

-1/21
3/2J

(c) 0.925.

□□□
l l • I f •!-

•.• { i ' .

.EIGENVALUES AND
EIGENVECTORS • 1 r-

• I

I • f r- - •

,.. ,
i
' ... ~·· I • • I -I . .. \.

LEARNING OUTCOMES
.I

After studying tlds cllapter you should be able -·to:

& Elgen Value and Eigen ~ectors
& The Characteristic Equ~t1on
} .
'l!I. Dlagonallzatlond Linear Transformation .. ....,
a Bgenvectors an \ : J

a Complex Eigenvalues
a Discrete Dynamical Systems
 11:A'J'DIIATIC8-II

. ...
d t x where A IS n><n matrix and
.
The goal of this chapter is to show how Ax IS ~elate ; x 'is non-zero vector in R2 SU xis a
column vector in R". For example, if A is 2><2 matnX and . ch ~t
. thr h ori'""
0
deternuned by x gets
Ax = AX for some scalar ;l.., then each vector on the line oug - - II\apPeci
back on to the same line under the multiplication by matrix A.

7.1 EIGEN VALIJE AND EIGEN VE£TOBS

I Example 1: Is[~] an eigen v!ctor

Solution:

f3
Since, A =LB [1]
of[; . _o] ?
OJ and x = 2 . So,
-1
1

Ax= [3 8 -1 2 - 6 3 [1]
01[1]-[3] 2 3x. = =

Hence, x =[~]is eigen v~tor ot[3 OJ

8 -1 ·
3 • vector of [1
Example2: Isxc[-2J•is eigen
~?
Solution: 5

Since, A= Lsfl 6
3]
2J1and x --[_2 , so

Ax= fl
3 Ls
61[ 3] r-9J [
2J -2 = 11 .., A -~J.
Hence, - -2x-[ ]· not eigen vector of[l
IS 61
5 · 2J·
.- .
Example 3: Show that -2 . . ['
Solution: JS e1gen value of 3 -13] .
Given, A =-2 and A=[7 3]
If 3 -1 ·
Ax=A,x
or, Ax =-2.t
or, (A + 2I)x = 0
...... (i)
 Eige~lllld~
non-tri"vial . lu tion, then A== -2 is ·
80
[7 ~~ ?)
- ........_.
hSS eigen Value of 3]
[
5ince, A + 21 = 3
7 3]
-1 + [~ 1H: n. 3 -1 .

So, row reduced the augmented lllat:rix is

[A+ 2I O]

........... -[: 3
1 gJ
[3 1
h . ~3 , } gJ
J

' . -~ ~.[~ . ~ ~].

Thus homogeneous system has free variable (here x2 is free variable), so equation (i) has non-
trivial solution. Thus A. = -2 is eigenvalue of given matrix A:
I'

For c;orresponding eigenvectors;

Th~ gei\era1 solution is formT;~ = [-:~ 3

] = x2 [ ..
1 3
{ ] ·
,1 '
I- ;

So, .r2 [-1{ 3] , where X2 ':I: 0 are the eigenvectors corresponds to eigenvalue A. = -2.

Example 4: Is 5 an eigenvalue of A = ~ [6·

-3_
1]
i ~ ?, 1
1

!. :_
Solution:

Here A = 5 andA -[~

- -2036~]-
2

H Ax=AX
or, Ax=Sx
I '
or, (A - SI)x = 0 ........ (i)

.
I ; [6
3
'vial solution, then A = 5 is eigen value of 2
I r

has non-trt

Here,
-3 0

A-5l=H 0
2 iJ - rn
5
0 u
-3
=[i -5
2 ~].
 JIATSSJIATlcs,-11 51 O]
So, ,ow ,educe augrnenred inatriX iS {.A-
-3 1 '-

-H -5
2
5
1 !1
-3 1

-[i 4
8
2
-1 !1
-3 1

-[i
n,us the homogeneous system
0
4 2
!1-
-5 haS no free variable, So, the equation (i) has trivial solution,

which5·means
Example Find )..the
== 5 is not eigenvalue
basis of A.
for the eigenspace corre•PondinS to listed eigen,value, where

. A==[!1 i !31 and 1 1'. = 3.

2 4 9 J Ax =.3x has non-trivial solution.
Solution:
Since A == 3 is eigen value for given matriX A, so
i.e. (A - 3I)x = 0 ........ (i)

has non-trivial solution.

Here,

A - 31 = -1
[
4 2 3] (3 1 -3 - O
o3 o~
0 =
O 3 .
li~
-1
2
2 4 9 O
O]
So, reduce augmented matrix is [A -31

=[~ ; } g}
-[i ~ i ~]
Also, neous system has non-trivi·a1 solution,
. . because x2 and X3 are free variable.
Thus the homoge
'

Xi + 2x2 + 3X3 = Q
X2 is free

X3 is free.
This implies,
X1 = -2x2 - 3X3

X2 = free
 X3 = free

is eigen space d basis

an for eigenspace is

rroof: Consider 3><3 upper triangular ntatrix entries Pl:I its

~
JV,
-• . Ax !:: AX
A and A be eigenvalue of A..

i.e. (A - Al)x = 0
re has non tpvial solution. ········ m -
( •,. • • I
Here,

[aug
'' .
., .' .
au
au] [A 0

~
A-Al= a22 a23 - O "-
0 clJ3 0 0

=
[au0-A a12
a22-A a23
au ].
0 0 cl33 - A
So its augmented matrix is ' • f ,... l l •• I
., t j t

au-A
= 0
[
0

e eq~tions (i) has non-trivial solution if£ (if and only if) in augmented matrix of A - ll
one of enbies au-A, a22-X, cl:33-A, is zero. This will happen only when'A. equal to one of the entries
1

au, au, a13 in A. Hence, eigenvalue of triangular matrix A are on the entries of the main diagonal.
.
Example 6: Find the eigenvalue of matrix O
[4 -2 1
· 0 0
Solution:
• Since, given matrix is upper triangular matrix, so eigenvalue of given matrix are 4, 1 and 6.

• 1e {Vt, V2,
Proo£: If poss1b . , , , · ., Vr} is linearly dependent. Since Vt ~ 0, so one of the vectors in that set
is a linear combination of preceding linearly independent vectors~ , · ·, ,
. ·linear
Let Vp+t IS . . . of linearly independent vectors vi, v2, .. • ••., Vp, Thus,
combination

Vp+t = C1V1 + C2V2 + •·· •·· ·· • + CpVp , . .'....... (l)

 JiATSSJIIATICS-U
Multiplying both side of (1) by A, we get
Avp+t = c1Av 1 + c2Av2 + ··· + cpAVp
¥• vi", s c,).,v, + c,).,v2 + .. . + c,),,,v, ········· (2)

1,ecause A vi = A.Vi for all i.

Multiplying both side of (1) by A.p+t we get 3 1·
1
¥1 ""' = c,¥1V1 + c,Ap+1V2 + .. . + c,¥ Vp •••••• ( )

Subtracting from (2) to (3), we get s

Ct().1 - Ap+1)Vt + ... + ep{Ap- Ap+t) Vp = 0 ........ (4)
Since (v , v~ .. . , v,1 is linearly independent, so c1(A,-A,,+1), c,(N.! -Ai,+1), ·· · . - Ap,1) all iQ~
c,(>.,,
bee all 9
1 (A, - Ap+,), (A-2 - Ap+ ), ... (1..,, - A,,+1) are not zero
zero. But ause eigenvalue A;
. 1 be iQ~

tinct, so c1, c2, ..., cp must zero.

dis from (1), ""' = c v + c,v, + . .. + c,v, = 0, which is not possible because eigenvect
Thus
11
are non-zero vectors, hence our supposition is wrong, so {v,, v2, ... , v,I is lin °" 1
independent. early

J
8

Example 7: Find the eigenvalue of [ : : - 6 ]·

0 0 2

'
Solution;
Let Abe the eignvalue of th · ·
/ - l. _ : :ven matnx. ~en characteristics equation is
3 6
68

10 0 2-1 : 1 ; ; ,. , r

., ' -'
(3 - ~) / -:-- ,.. 6· / = o·
O· 2-A ., f . '

A= 3, A= 0, A= 2

..i_ EXERCISE 7. I
1~.Le=tA=;fl~61~[6~] --;::----'--~-:-----~ :J
Ls 2J , ~ = -5 . Is u is eigenvector of A?
2. ,Is[[} an.eigenvector,of [-3-3 81] · ? •

3. Is [fl an eigenvector of [23 1 ]

4. .· of[3-; -1-:/ 19]

1s[!]1 an.eigenvector
4 ?lfso, find
· the eigenvalue.
 15 A. == 1 an eigenvalue o(matriJc [2 2]
5, - 3 5 _?
A5
15 = an eigenvalue of Dlatrix [5 OJ
6. 2 1 ? If so find
1 2 2] one correspondin .
7,
. ,
[
Is i = 3 eigenvalue of 3 . _
0
1
2
1
'
1 ?Usofind
one conespondin .
g eigenvector.

g eigenvector.

tj}~
8. Show that 2 is eigenvalue f [
4
-1
o Dlatrix 2 I 6
6]
, · · 2 -1 8 . ,
~~ 9. If I\, IS eigenvalue of invertible ma . .
of A-1 and x is corres din . bix A, correspondin .
pon g eigenvector. g eigenvector x then 1/'A. is an eigenvalue
~tors
early Hint: Ax= AX =>A·tAx = M-i x so we get A•tx = ( _1)
A. X
10. Find a .basis for the eigenspace corres .
5 07 ponding to listed eigenvalue
a. A =[ 2 1J and A =1. b A=
·
[10 -9J
4 _2 and ).. = 4.

c. A=[_~ ~2] and"-= 10. d. A =[224 ~1

-1
:J
8
and)..= 2.
1

~ A=[r ~ tJandA=l andA=2.

11. Fmd the eigenvalue of following matrices

a. [g0 0~ -1~] . .
b. [~ · ~ . ~] .
1 0 ' 3
12.. H x is an eigenyector for matrix A correspond to eigenvalue 11., what is A3x?

13. H x = (_~) is an eigeiivect~r for ~tiix A = [~ ~]'correspond to ~ ~ -f Find A3x?

2. No, Au*Au. 3. No, Au* ).u. ,

4. Yes, Au= O.u, 0 5. No. 6. 1 °Yes, ~t] ,·k e R- {O} . '.

' .

YesifJ
l I •

7. k e R- {OJ
10. a. {[~]} b. {[~]} . : ~

I t

c. {[t]} d.
mJffJ r e. For~=4 {[1],[!] },forA=l, {[-n}
b. 11. = 4, 0, 3 C. A.= 3, 2, 1
11. a. A.= 0, 2, -1

12 A.3x
13. [-384]
320
- l
 11.ATIIDIAflc&-a

7.2 THE CIIABACTERISl'IC EQIJATION

on
(A-Al) is
ol the matrix A.
. . po1yn omial
&ample 1: Find the characteristic of matrix[; -1J
4 • Also find its eigenvalue.
I

Solution:

Otaractenstic polynomial is IA -U I,
where

A -Al = 1 [2 -11
4 - LO 01 = [2-}..
p.. }j 1 -11
4-A .
Therefore, characteristic polynomial is,

12-}..
l
-1
4-}..
I
= (2-}..) (4-A.) + 1
= }..2 - 6}.. + 9.
So, characteristic equation is I A - Al I = 0.
or, A.2 - 6}.. + 9 = 0.
or, A.2 - 3}.. - 3.l + 9 = 0.
or, A(A - 3) - 3(.l - 3) =0.
or (A - 3) (A - 3) = 0.
or, A =3.

Therefore, A = 3 is eigenvalue of matrix[_; ~].

&ample 2: Find characteristics equati~n and eigenvalue of A where A =
Solution:
[! i] .
Given,

A=[i i].
So, the c:haractenstic equation of A is / A -
Here,
Al/ = 0.

A-Al={: il-[~
Thus, characteristic equation of A is,
f/=[1/ 2:4].
/A-Ai/ =O.
OF /1/ 2~/ =0.
or, (1 - .l) (2 - .l) + 16 ::: 0.
 Blcewname and Elpuveetor w l§iinajl
or, A. 2 - 3).. + 18 = 0.
'fhiS gives,
3 + ✓ (-3) 2 -4 X 1 X 18
A. 2

3+£63
= 2

,r1..:., oives
u...., o- the imagm·ary value of '\"-· Therefore, the matrix
· A has no real eigenvalue.
·

·ces. The matrix ~ i& similar to matrix

t P-1AP B, or: ·vatently A • PIJp.. \~ Ud , ...
:A an ..
tlon.

rroof: Since A and Bare similar so there exist invertible matrix P such that B = P-1 AP. Then we wish
to show

where I be an identity matrix.

Here,
B - ll = P-1AP - ,J>-1P.
= p-1AP-P-1llP
Then, IB - A.I I = IP-1 (A - A.I} P I
= IP-1 I IA - Al I IP I

= IP-1 I IP I I A - A.I I

= IF-1PI •IA-All
= 111 IA-ill
= IA-"-1I.
Therefore, ._, ·J
Characteristics polynomial of B = Characteristics polynomial of A.

Hence A and B have same eigenvalue, because

IB-lll = IA-ill
=O
gives the same value of 1'..
 IIATJIBIIATIC&-11 ,_:1 then show that det(A) = det(B).
A d B are sm war,
Example 3: If n><n matrices an
1
Solution: . . ertible matriX p such that B = P- AP.
,_:1 there extst mv
Since A and B are swwar, so
So, Bl = IP-tAPI
==' I P-1 I I A I I P I

== I P-1 I P I I A I
== I p-1p I I A I == I I I I A I = I A I -

Thus, IBI = I A 1-
Example 4: The square matrix A = (
-l
3
2)
1 an
dB= ( -56 -: ) are similar to each othet

~ ).
2
through the invertible matrix P = ( _ 1
,C
Solution:
To prove the matrices are similar, we must find the inverse of matrix P.

I Then,

p-1 AP. -(0 -1) · c-13 12) · ( -12 1)

- 1 2 0
I ,

=(-/ -: )
=B So,
Application to Dynamical Systems

f 0.95
Example 5: Let A = LO.OS
0.03] .
0_97 . Analyze the long-term behavior of a dynamical system defined

by Xk+1 =Axk for k =0, 1, 2, .... with Xo = [~::J.

Solution:
Let's find eigen values of A and a basis for each eigen space.
So, the characteristic equation for A is
/A-ill =O • I

0.95 - }i. 0.03 1-

I 0.05 0.97 - }i. - O
(0.95 - }i.) (0.97 -A) - (0.03) (0.05) = 0
or "-2 - 1.92 A+ 0.92 = 0
 2
1.92 ± )f(l.92) - 4 >< 1 >< 0.9~
.,, - 2 >< 1
~

1 _92 ± ✓o.0064
.,,- 2

1.92±0.08
== 2

~ == 1, 0.92
for eigenvector, A.= 1,
[A-A.I 0]
-0.05 o.o3 o]
,::: [ 0.05 -0.03 0
[-~05 oi3 ~]
x2 is free variable so

v1 = [s·3].IS eigenvector
. corresponding to eigenvalue)..= 1.

similarly, v2 = [ _;] is eigen vector corresponding to eigenvalue A= 0.92.

{vi, v 2} is basis for eigen space, since :xo = 0.4 .

I
[0.6] .
•

So, Xo = Ct Vt + C2 V2 = [v1 v2] [~~

• ' a 1 I ' •
..
Ct1
i.e. [ cJ = [vt v2J-t Xo
[3 17-1 [0.6]
= 5 -lJ 0.4 r

= _!_ [-1 -lJ [0.6]

i .

-8 -5 3 0.4
I

-[0.125]
- 0.225

Ct = 0.125, C2 = 0.225

Now,
Xt =A'4J =A (ct Vt + c2 v2)
= Ct (Av1) + c2 (Av2)

x2=Axt = A (c1v1 + 0.92 c2v2)

= CtVt + (Q.92) 2 C2V2
 IIATHEIIATICS-D

Xk = Ct Vt + (Q.92)k C2 V2
1
1
Xk = 0.125 [;] + (0.92)k (Q.225) [ _1] .... (1)
. f th difference equation
This explicit formula for Xk gives the solution o e
0.375]
Xk+t = Axk. Ask ➔ oo, (0.92)k ➔ 0 and Xk ➔ [ 0.625

[~:!~] = 0.125 [;] from equation (1).

Eigenvectors and Difference Equation

H there is a matrix A of order nxn such that

= Axo,

I
X1
X2 =Axt _
and in general,
Xk+l = Axk ..... : (1) ·

I foi: k = 0, 1, 2, ..., is called the linear diff~rence equation.of sequence {xn} in Rn. :ro compu
xo is, lqlq~1;,,,$,t,:!fai?'.; c;,;:~~-~·,;i,:~ih ; '~+,;i;,,.;;r,r · ..~ ,, . ,
X1, '.%2, X3, and $Q 911, ,provide.d _

And, the solution of (1) is, fork= 1, 2, ...

...... (2)
where A is eigenvalue of eigenvector xo, because
_ ,, P, • • 11
Ax1c = A(Akxo) = AkAxo = Ak(AXo) = (AkA)xo = J..k+1xo = X1c+1.

Example 6: Let u and v be eigen vectors of a mabix A with corresponding eigenvalue A and µ, and
let c1 and c2 be scalars. Define,
Xk = CtA.k u + ciµ_kv (k = o, 1, 2, ....... )
a. What is X1c+1 by definition?
b. Compute Ax1c from the formula for x1c, and show that Axk = Xk+i.
Solution:
a. Since, x1c = c1Aku + c2J!"v fork= O, 1, 2, ...
Thus, X1c+1 = c1A1c•1u + c2µ1c• 1v
b. Ax1c = A(c1Aku + c2Jlkv)
= C1AkAu + c~kAv
= CtA.kA.u + c~kµv
Since, Aand µ are eigenvalues of u and v respectively.
= CtA.k+lu + c~k+lv
Thus, Ax1c = C11i.k+lu + c~k+lv
Using (a), we get,
Ax1c = Xk+l,
 find the characteristics polynomial d .
1- an eigenvalue of

a. [~ g b. [ ! 1]

· [I !2 =fl
g. [i ; i] h. [=} ~ l]
1 0 0 0 0
0 2 0 0 0
0 0 -2 0 .0
j.
0 0 0 3 0
0 0 0 0
2. Show that if A = QR with Q invertiql~-then A is similar to A1 = RQ.
Hint: Here, Q-tAQ = Q-t(QR)Q = (Q-tQ)RQ = RQ = A1, so A and ~ 1 are similar.
3. What do you mean by eigen values, eigen vectors and characteristics polynomial of a matrix?
Explain with suitable example.
0.6 0.3] [3/71 [0.5]
4. Let A = [ 0.4 0.7 , Vt = 4/7J, Xo = 0.5
a. Find a basis for R2 consisting of Vt and another eigenvector v2 of A.

. µ, and b. Verify that Xo may be written in the form Xo = v1 + cv2.

=1 2 d fin x = Ak Xo. Compute Xt and x2, and write a formula for x1ir.. Then show
c. For k , , ... e e k
that X1t decreases as k increases.
,,, '::.~re.~~~
_'fi.,'d~
 IIATll&IIATIC&-11

ntfon Tlieorem .
aiagonaliz.able iff Ahas n linearly indepe.t!dent e i ~

gunalizing a Mabix
- A<Xbldat,t to diagonalization theorem, to diagonalize matrix A, steps are t:
.,.81:ep i: Fmd n linearly independent eigenvectors of A say v1, v2, •· ., Vn. · • ' ~
~ , 2 : For matrix P having vi, v2, ... , v n as its column vectors.
Step S: The matrix D will be the diagonal ~trix with Ai, A.2, ... , An as its successive diag
entries, where A.I is the eigenvalue corresponding t~ ~i for i =·1, 2, ... , n.
Here, A = PDP-1 or AP = PD, if so our p and D r~aj)y :'York and A is piagonalizible.

i i-1] ,
2
Example 7: Diagonalize the matrix [ 3 r.
if possible.
-1 -2
Solution:

Let, C,

A=U-l .
2
3 -1]
-1 .
-2 2 o I .. ' • 1

Here, (

A-AI =U-1 3
2

-2
-lJ
-1 - 0
2 · 0
r ~] 0
)..
0

= r-A
1
2
3-A -1]
-1 >.
-1 -2 2-A
And, the characteristic polynomial of A is - I

/2/ -1
2
3-A
-2
-1
-1 =O
2-A.

- rA 1
0
2
3 - A,
1-l
-1
-1
1-l
=O.
 2- A. 2
- (1- l) ~ 3 - A.
1 1

Either)..= 1, or
2-~A.
4-A.
0
-11
3
-11 =0.
\

⇒ \2;A 4:A.\ =0.

⇒ (2 - A)(4 - A) - 3 = 0.
⇒ A2 - 6A + 5 = 0.
⇒ ).. a 1, 5.
~us A= 1 and A= 5.

For ).. = 1, since Ax = A.X.

so, (A - ll)x = 0.

And its augmented matrix is [A - Al O].

= [A-I O] _ [Since ).. = 1.]

1 2
= 1 2
[ -1 -2
1 2 -1 0~
~ 0 0 0 0 .
[0 0 0 0

Here, x2 and X3 are free variables and x1 is basic.

From the last matrix we have,

Xt + 2%2 - %3 =0

X2 = free

X3 = free.
r . ,[ Xt] [-2.xl + ~3'] { (~21 'r.:YL I

So,
"= ~: = ~: 0 %2~1f "'~~Jj.
ding to)..= 1).
Thus bases for eigen space {corresPon

and v2=Hl.
 IIATBEIIATICS-D

Augmented matrix [A-U OJ

= [A-51 OJ
-3 2 -1
=[ 1 -2 -1
-1 -2 -3

+~ -1

-2
-2
2
-2
-1
-1
-3 i]
i]
-1

-rn -4
-4
-4
-4

0 1

-rn 1
0
1
0 il
Here, x1 and x 2 are basic and X3 is free variable.

I So from the last mabix we have,

X1 + X3 = 0.

X3= free.
This gives,

,.
X3 = free.

Thus, basis for eigenspace (corresponding to A= 5) is v, = [ ~!] .

There are three bases vector in total, which are linearly independent eigenvectors and are

v, =n1- =mand v, =[~!J.

V2

So, P=
r~
1
0
1
-1]
~1 and D=
[1
~
0
1
0 n
''
 J\).50, -----W~We½#jl

[-2~ 1
-1][1 0
o]O • [-21 1
PD= 0
1
-1
1
0
0
1
0 S O
0 -5 .
1 -1
5

d APm[~-1
2
3 -1] [-2 1 -1] [-2
~
-1 1
0 -1 ... 1
t0
an -2 2 O
r •

1 1 0 1
This shows that AP = PD or equivalently A. PDP-t.
So, A is diagonalizable.

£xaDtple 8: Inagonalize

Solution:
~ matrix [ a ~ n foist

Let,
_,.
-1 4.-2]
A= -3
_3· [ 4
1·
0 .
3
[.I = n:
So, the characteristic polynomial of A is

-1-A. 4 -2] I.
A-Al= :-3 . 4-A O .
·. [ -3 1 3-A.
J

Therefore; the cliaracteristic equation of A is

A-Al= 0.
u
4 -2 lo
·-1.-A
⇒ -3 4-A 0 =O
-3 1 3-A r
-2 I 1 ,
-1-A 4 )
-3 4-A 0 =O.
⇒
0 -3 + A, 3-1 r
-1-A 4 -2
'(3- A) -3 4-A 0 =0.
⇒
0 -1 . l 0

-1-A 4 -2 IJ

4-A 0 =0.
Either (3 - A) = 0. or -3
-1 1
0

-1 ..:i 2 -2
4-A 0 =0. J -

⇒
-3 1
0 0
 IIATBBIIATlc&-11

⇒ 1-1_; l 4~A, ==0.

⇒ A.(-1-A.)(4-A.)+6==0.

⇒ A.2 - 3A. + 2 :: 0.

⇒ A.2 - 2A - A.+ 2 = 0.

⇒ A.(A. - 2) - l(A. - 2) = 0.

⇒ (A. - 2)(A. - 1) = 0
Therefore, A = 1, 2, 3.
For A"': 1,

Since Ax = AX. So,

(A-ll)x = 0
And, its augmented matrix is

[A-ll OJ
= [A-I OJ [being A.= l.] L

=
r=; 4
3
1
-2
0
2 iJ .,.

-u
-2 1

-3
-1
1
0
2
g] • . ~ • • .. •J C ~ .6"! L.fl

-2 1

-[i -5
1 -1
5 !]
-2 1

-[i 1
-1
-1
1 ~] ./\
- I

-2 1

-rn 1
0

0 -1
-1
0 u
-G 1
0
-1
0 i]
From the last matrix,
X3 is free variable
and X1 -X3 = 0.
X2-X3 = 0.
 This gives, .............. Blcc.c,

x1 =4 x2 =X3 and X3 =X3.

~ x~[:J-[:}{ x,m}.
Thus basis for eigenspace (corresponding to A• l) is v, • [n.
,I\

·
Similarly, •I • I
we get . basis for eigenspace ·(corresponding to A = 2) is v =
2
[2]! and basis for
l
eigenspace (corresponding to A. =3) is v 3 =
' • [1]: .
[Procedure is exactly same as above for A = 1]. - I i

c.m 1
, :..,· i bn

~ are three bases vectors in tota\ which ~ I

linearly1 inde~dent and are v1
, - [1]
= ~ ,
Iu I, u
n

n
I) l
2 0
So,
P=G 3
3 !] and
D=rn 2
0
) I
lj

Also, J.i j,: •j •I :

134
,[1 2 1] [1 0 OJ [1 3]
PD= 1 3 3 0 2 0 = 1
003 1
4
6
6
9 .
12
•• ")l.{'(l •trl.

and
. [-1 4 -2r 2~-[! !]Jo}
AD= -3 4 0 1 3
4
6
-31313 r 6 12 I

· alently, A •-PDP-1•
Thus, AP= PD or eqwv

Therefore, A is diagonalizable. ,J

Example 9: Diagonalizable the ma~

, . A = [:·
0
~O ·'s~] , if possible.
Solution:
I •w lJ

Giv~ • . j

n
,·. t.lJ

o·
4
A=rn 0
 So, the characteristic polynomial of A is

~
4
A-ll=[ ~A 4~).. ].
~
0 0 5-A.
Therefore, the characteristic equation of A is s
IA-All =O.
4-l 0 0 :..
⇒ ·1 4-l 0 =O. l
0 0 5-l
This determinant is an lower triangular. So, we get, I I
•,. 1

).. = 4, 5.
For)..= 4,
Since Ax • AX. So,
(A-ll)x = 0 '3 I
' t
And, its augmented matrix is
[A-ll 0)
! = [A- 41 r. OJ ·~ · · [being1: = 4) -,:,. ·1 •

I
,f
' I

0 0

-rn 0
0
0
1 i] ' '.
0 0

-rn 0
0
1
0 ~]
1bis is a reduced echelon form.
From this last matrix,
x2 is free variable
and X1 =O
X3=0

~re, = x [:J [!] x.[~].,..

=

[07
~J .
I .·

Thus basis for eigenspace (corresponding A= 4) is Vt = • l ' I •

Similarly, the basis for eigenspace (corresponding .l. = 5) is v, -rn1.

. There is only two vector in basis and is linearly independent. But we need linearly three
independent eigenvector to form P, so P is doesn't exist. Hence, A is not diagonalizable.
 Blpsnalae and ElpJneetor W \CBAfflll11
rro"f: If vi,_ v2,A,.· · ·th·., Vn bare eThigenvectors corresponding to the distinct eigenvalues )..1, A.2, ••..•• , A.rt of

~= J1L-3~r~1~J:::::~b:gonalizable.
rnatrtx
. nalizabl en th Y eorem. 2' v 1, V2, ••• .. .. , Vn are linearly
· ·
mdependent; hence by

Is ma~
2
solution: .
Since matrix is triangular and there are three distinct eigenvalues (i.e. ).. = 2, 3 and 5) and matrix
is 3><3. So it is diagonalizable.
th
~pie 11: If A • PDP-1 for some invertible matrix p and diagonal matrix D then prove at
Ak-=PDkP-1.

5otution:
Since A= PDP-1 . So,
Al = (POP-1)2 = (PDP-t) (PDP-t)
= PD(F-1P) OP-1
=PD2P-t

Again,
A3 = A; Al = (PDP-t) (PD2P-t)
= PD (P-tP) 02p-1

=PDID2P-1
=PIYP-1
- l I

So, in general, for k ~ 1,

Ak = pokp-1.

= PDP-' compute A'; if P "'[~

ExaJllple 12: Let A '

"t
. ..
' ' ' "\
 IIATJIEIIATICS-U

.Example 13: Compute A8, where A=[~ =f] •

Solution:
HA is diagonalizable, then A = PDP-1, so A8 = PD8 P-1 .•..• (i)

For P and D, we find eigenvalue of A.

Its characteristic equation is,
IA-UI =O.
I
4-A -3 I
⇒ 1 2 -1-A = O.
⇒ A2 - 3A + 2 = 0
This gives, Ai = 1, 2.
For A = 1, the augmented matrix of A is,
[A-U OJ
[A-I OJ [Being l = l.]
-3
=[~ -2 gJ
-1 :.
~[11 -1 gJ
-1
~ Ll0 0 gJ.
From the last matrix,
.r2 is free and x1 - x2 = 0.
Titis implies, .r2 = free, X1 = x2.
-:
Therefore,

Thus, the basis for eigenspace (corresponding A= 1) is v1 = [!].

., .
Similarly, for A = 2, basis for eig~nspace is ·v2 = [~] .
,. - I:..

Thus, there are two basis v~ors in toptl, which are linearly independent and are

v,=GJ and v,-m.

So,

From (i) we get,

AB = pI)Bp-1
 Elaenvalue and Blpnvector w -=---
~ [! m~ TI'(l)L~ 7J
=[1 3][1 OJ[-2 3]
1 2 0 28 1 -1
8

=[~ m~ ~t
2
· !J

-[1 7687[-2 3]
- l . 512.J 1 -1
766 -765]
• = [ sio -509 ·
'
sxample 14: Let A ~ [~ 1;], Vt •[~and v1 =· [!] •rS"°ppose you are ~Id ~hat v1 and vi are
. , eigenvecto~ of A. Use~ infj rmation to dia~palize A.,..
Solution: , · '_r
~ '
To diagonalize A, we mustrfind the value of P and p. For these, we need the eigenvalue A. of A.
l . I • ' . I . f -[3] I :..

Foi the eigenval~e ),. c~rresponding to eigen~ect~r ~ = 1 .

("'

Let, 1-
... l ,::.

Av,=[=~ 1i][n=t!:1i]Jrn=.~[P= 1"' 1 . ,

, ~ l < I I • , . -~ ', . .J b,,, i I •

This shows that X =~·:· ,✓ "

For the eigenvalue A correspon~g to eigen~ector vi= [iJ.

Let, r
I

_, I' >i
A
This shows that 1 = 3. !: '. .
f \
. So, l I

,
! t 'l I 't
And,

..!

· alently A= PDP- -
1 ( .
This shows AP = PD or eqUlV .

So, A is diagonalizable.
 IIATBEIIATICS-11

:.:I_HX~RCISE 7 .3
1. Let A = PDP-1 compure AS, where p = [~ i] and D = [~ ~] ·

2 HA = [ : ~] compure A4• Also evaluare An, where n is positive integer.

3. Determine if the following matrix are diagonalizable ~

a. [~ ;
. 0 0 -3
!] b
. 1 2 2
[! ~ ~]
·1 .
c
.
[-; ~2
4 -. 6
~l~J..-
4. Diagonalize ~ followittg matrix, ifpossj.gle - .
.... 501

a. [! -~l !l0 ,OJ

0 0
-3 0
1
0 -3

5.

1.
-92 372]
A5 • [ -31 125
[ 1169 544 J [
2 X Sn - 3n Sn - 3n 7
2· A'= .::. 4088 .,;.; 463 ' (V_' ~. -:-2 X 54 + 2 X 3n- - Sn + 2 X 3nj
3. a. Since matrix is. triangular with 3 <f!sti!tct
ef~va\_ue A=;: 2, 73~4_~d is 3x3, so diagonali7.able.
b. Since matrix is triangular with 3 distinct eigenvalue A= -1, 2, 3 and is 3x3, so diagonali7.able. E
c. Since matrix is triangular with 3 distinct eigenvalue A= 1, -2, -3 and is 3x3, so diagonali7.able. ,. ·
d. Since matrix is triangular with 3 distinct eigenvalue A= 0, 4,-5 and is 3x3, so diagonali7.able.

4. a. p = [! ~], D = [~ -~J b. ~
P [-: !] ,D = [-; ~]

c. Not diagonalizable. ci. P=[i ~o ~2];'o ='[~ooiJ

1 1
.• ~ ~l
e. Not diagonaliz.able. f. No~diag~nalizable. ~ g. , P {:1 -: -~] , o = [~ ~2 -~ l
, .1 0 1 ' ' O ·O J
o , D [:O !O 5~ ~
0 -8

h. ps[! 0
0
1
4
1
0
:.i6]
1
= O
0 0 0 5
i. N of diagonaliz.able. ;
r I
.(

~],o=[~ I ~ ~l~
3 -1

j. P•[~ 2
0
0
-1
1
0 1 0 0 0
1• I

5. p= 2 [1 OltJ and D- =[S.0 -1OJ · , I

 Blpmalae IIIUl Bi&enuecitua: W ICBAPmt'11
f.4 £iG£NVEcroRS AND 1ANEAa TRANSFORMATION
i;:;E?- -~~
~

Solution:
Given that
T(bi) = 3d1 - 5d2
Therefore, [I(b,)]o = [!l ',,.- .. , . "'·'. '·,
Similarly,[I(b,)]o=[·:]an\i['Th,)lo"[~l • 0~ ' : _ ; , / ' . • _ : ·,, .,: ; , ,

Thus, matrix for T relative to B and D is,

[T]s.o = [[T(bi)]o [T(h?)]o· [T(\,3)0]
..}

=[1 -: ~]. ·\ I
-I ..
' It

Remarks: Let T: V--+V be the linear transfo'11"1-tj.o~ ,with ~_13 7 {b,, b,, .•., bnl of'vector_space V.
· Then matrix for T relative to B or B-matrix for T is,
[T]s = ([T(bi)]a (T(l'2)]s ... [T{bn)]a], , ~ 1'
➔&
1
Example 1: Find the matrix representation of linear transformation T: R. defined by
T(x, y) = (x, x + 2y) relative to basis (1, 0) and (1, 1).

Solution:
The matrix for T relative to basis B = {(1, 0), (1, 1)} is
[T]e = [(T{bt)]e [T{ln)]e).

For [T{bt)]e I •
'\

Since, T{bt) = T(l, 0) = (1, 1) = ct(l, 0) + c;z(},,1L - '

= (ct + c2, c2).

(1, 1) = (ct + c2, c2),

Equating correspanding entries,
...... (1)
ct+c2=1
...... (2)
and C2 =1
This gives, Ct= 0.
 MATBBIIATlcs-D

Therefore, [T{bi)]s =[~] ·

For [T(b2)]e
T(b2) =T(l, 1) = (1, 3) = c1(l, 0) + c2(l, 1)
=(Ct+ C2, C2)
⇒ (1, 3) = (Ct + C2, C2).
Equating corresponding entries,
Ct+·c2=l ....... (3)
c2=3.
This gives, by (3), Ct = -2.
. [-2]
So, [T(bl)]e =
3
.

Thus, [T]e = [ 1o -2] .

3
Example 2: Let T: P2 ➔ P3 be the linear transformation defined by
T(Pct>) = (t + 5) p(t)
(i) Find the image of p(t) = 2 - t + t2
(ii) Find the matrix T relative to bases (1, t, t2Jand {1) t, t2, !3},
Solution:
(i) Given T(p(t)) = (t + 5) p(t)
So, T(2 - t + t 2) = (t + 5) (2 - t + t2)
= 10 - 3t + 4t2 + t3.
(ii) Given B = {l, t, t2} and C = {l, t, t2, t3}.
So, [T]s. c = [[T(l)]c [T(t)]c [T(t2)]c).
For [T(l)]c,
Since, T(l) = (t + 5) 1 = t + 5 = 5 + 1.t'+ 0.t2 + O.'t3'

⇒ IT(l)]c ~ [!J ".

For [T(t)]c
T(t) = (t + S)t = t 2 +St= 0 +St+ 1.t2 + O.t3

⇒
t ,J r

IT(t)]c=[fl
And, for [T(t2)]c;
T(t2) == (t + 5)t2 = t3 + 5t2 = 0 + Ot + 5t2 1+ lt3

⇒ ~W)lc=m.
Hence,. . .[T]e.c = [! l F
0 0;J
 ~a,np1e·3: Let T: p 2 ➔ p be ~
> the line
' 2 811d Eige~
. T(p(t)) == p(3t-S). . ar O"ansforznation defined m e#fi-]
Find fTJe With . by
Solution: respect to basis B -1
. - 1, t, t2J.
Since T(p(t)) == P(3t _ S).
That is,

T(ao + a1t + a2t2) == ao +

We know that a1(3t-S) + a2(3t-S)2.

fTJe = ffT(l)Je CT(t)Je 2

So, for fT(l)]e, CT(t )J 8J

T(l) = T(l + Ot + 0t2) == 1 +

0(3t - S) + 0(3t - 5)2 == 1 ,-,
== 1 + 0.t + O.t2

Therefore,
ff(l)Je = [ff
For fT(t)]e, ,,
...
,.
..,,.

T(t) = T(O + 1.t + O.t2) 1.::: · • I-

.._
., - ~:..:~ 0
r•r~ ~ 1, -:=_ =-+ 1(3~~ S) _+ ~(~t-,s12 .· . ,. - r
=3t-5 . I I I -- . I
. .

l .J l ,
Therefore,
r. . .
And, for {T(t2)Js,
•I • ..., ►

T(t2) = 'f(O + O.t +ft2j I'

J • I

= 0 + 0 (3t - 5) + 1(3t - 5)2

= 9t2 - 30t + 25 = 25 - 30t + 9t2.

Therefore, /T(t2)Je =[f].

1
Hence, [1J8 = 0
-5
3
25]
-30 .
0 0 9

By the definition of similarity, matrix D is similar to A. That is, B-matrix for the
transformation x ➔ Ax is similar to A.
 MATHEMATICS-D

Example 4: Let A = [! ::J, bi = [~], h2 = [~] and B . = (bu , h2). Find the B-matrix for the
transformation x➔Ax with P = (bi h2]. ·
Solution:
1
We know that by diagonal matrix representation theorem,
B-matrix for transformation x➔Ax is D= P- AP.
1

Since,

So,

I,
p-t =.!.[1 -2J =[·12 -32] ·
-1 -2 3 ·
,, ~
r r .,

Therefore,
f
,
I

=P-tAP=[·l2-34-821
2][4 -9][3 2] .....
- . ,.,,.,.
[T]e = D a ,,,.
=[·12 -32][-6-4 -1]0 [·20 -21] =
- '11/,jJ

,· • j ➔ f f, . - •, I
is required B-matrix for transformation ~~Ax.
1
[Note that in above example, matrix D is similar to A. So if thJ question is; fin d the similar matrix to A with
same infonnation solution is exactly same.]
Example 5: Let T: R 2 ➔ R 2 defined by T(x) = Ax be the linear traitsforination. Find a basis B for

R 2 with property that [T]e is diagonal, where A=[! !J.

Solution: L
We know that by Diagonal Matrix Representation Theorem,
· D = [TJ8 = P-1AP; where, a·= {bi, bi} be the basis for R 2• So that P = [bi °b2].

Since A =L~ ~. r. . ':l

I
So, characteristic-equation is,
IA-lll =O ' I

.'
⇒ -A 4-A
-3 1 ' =O

⇒ i2 - 4A + 3 = 0.
'
⇒ A= 1, 3.
For A= 1,
• I .•

The augmented matrix is,

.
[A-ll
J' O] =[-1 1
-3 3
 So,
and,
This implies ''
x1 = x2 and . , .. ,
X2 = free.
Thus x = [x l =
xJ
1
[1]1 X2.
I.
. )

Therefore, eigenv~ ·.
flSbi= 1 .r Il]
Again, for A = 3 · .. 1
< , ,i_ I
Augmented matrix'is r. . k l r -
~ !)J ~
'
1 • '( •
J , ••• I 1~ ' f l •I

=[-3- 1 . ojlL'
• , t

[A-AI OJ r .. iJ-
. ''
, . ;: , -3 1 0 • I

• • • I ,• • L ) .., ~[""3 "'1 : 07

!.•.I ~ ... TJ I ' rh
o o oJ· .i· : = l- ~
So, j n ,
' ' . ·,· ' ...
•
i . ~2 is free variable
. ,
And - 3Xt + X2 = 0. . .
This.lDl.pIi·es,.
1
r I - ·. l
• I

. X1, =3~2 and X2 = free. ' ,

-fi '.,; ~ ~ · ' :: r; -. '":
Therefore, x=[x11=[1!3].
xiJ 1
-~-3-,X\_3fl].r . X2 -
-! 1. • t ,

. •.l; .
[ •• _.;; • I
i : .l,.

I.. • -

i { J; I/
or ; ' r

J1ierefor~, -eigenvector is.bi ~-~;]

b, : ._· ),1 ,.J;=- . . • :i..' L-
I • r
• I

Henc~, required basis for R2, B =

.
[n, [!] }. . -,
' . . ,._
..

l~·_EX_.E_n_c1_s_1~_-1_·. 4 - - ~· r:......____:_______!_:,_
i\, __:__ _:_______:_~ ~'

1. Suppose B = {bi, 1'2} is a basis for vector space V ~d C,= {c1, c2, c3} is a basis fo; vector space w.
Let T: V ➔ W be a linear transformation with the, prp~"'rty that T(bi) = 3ci - -2c; + 5c3 and
T0>2) = 4c1 + 7c2 - c3• Find the matrix for T relative to B ~ C. ''

2. Let D = {di, d } and B = {bi, 1'2} be bases for vector $paces V and W respectively. Let T: V➔W be
2
a linear transformation with property that T(d1) =. ~~ - 3"2 and T.(d2) ,= -4bi + ~- Find the
matrix for T relative to D and B. : · -
3· Let T: R2 ➔ R3 be the linear transformation defined by T(x, y) = (y, -5.x + 13y, -7x + 16y). Find
2
the matrix , for the transformatio~ T with -res~t · b ~ l3 = {\b, bi} for R 1 and 1

C • {c1, c2, CJ} for R3, where bi= (3, 1), l'2 = (5, 2), c1 =(1, 0, -1), c2 =(-1, 2, 2) and C3 = (0, 1, 2).
 MATBEMATICS-U r-

4.
- T
Let T: R4 R 2 be the linear transformation defined by Y
= (x) c-y)
+Y
and let B == {1..
UJ.,
1...
"'2} be the

basis for which bi=G] and~= [-;J. Find [T]e (i.e. B-matriX for T).
5. Let T: define by
R4 R' linear transformation t(;) = Gx\~) and B ~ {bi, b,} be lhe basi,
the

for which bi = G] and ~ = [ i]. Find [T]e (i.e. B-matrlx for T).

6. The mapping T: P2 ➔ P2 defined by T(ao + a1t + a2t2) = a1 + 2a2t is linear ~~formation. Find
B-matrix for T, where B = {1, t, t2} is basis for P2.
7· Let T: P2 ➔ P4 be the linear transformation that maps a polynomial p(t) into tbe polynotnial
p(t) + t2p(t). Find the image of p(t) = 2 _ t + t2. Also find the matrix f~r T relative to the bases
{1, t, t 2} and {l, t, t2, t3, t4}. .

L
I
8. Assuming that T: P2 ➔ P 2 defined by T(ao + a1t + ~2t2) = 3ao ~ (Sao - 2a1)t + (4a1 + a2)t2 is linear.
r Find the matrix [T)e, where B = {1, t, t2}. ~v ·1

Let T: R 2 ➔ R 2 defined by T(x) =Ax.Find basis B for R2 with property

' 4 • r'

9. [T]e. ,is. diagonal, where

.

(i) A=[_; J] (ii) A= [_i -J]

10. Let A =[_~ -~J, bi =[ _~], ~ =[~]and B = {~~ ~} ,i,e the ~as~ f<?r R 2. ~in~ fue B-matrix for the
- I I

transformation x➔Ax with P = [bi ~]. · L; ~ · L ! "' ·

11. Let A = c~ !],

x➔Ax with P = [bt ~].
bt = [~], ~ = [-;] and B = {bt, ~}. Finq tl!e ~a:matmc for the~transformation
:: ~ : ·'.

. -· .

1. rllac=[-15 -1;] 2. [T]o.e =[_; i] 3. ·[T]e.c = 0

-2 -1
1 4. [!le =I; -~J
,r

5.
.[T]e =[30 ·-2]
-1
[O 1 OJ2
6. - [T] 8 '= 0 0 ·
0 .0 0 -

,r, . . .. .
, 1 0 0
1

. (f 1 0
<

7. (i) T(2 - t + t2) =2 - t + 3t2 - t3 + t4,

l

[T]e.c = 1 0 1 o ol0
-2
0 1 0
0 0 1
8.
rll•=U 4 1

~- (i) bi ; [;] ; bi = [-;] . (ii) bi =[ 1], b, =[!] 10. [~ ~] -. 11. [~ il

 Eigenvalue and Elpnvector W 1Ceijm'11
•
f .5 COMPLEX EIGENVALIJES

A non-zero x e ~ is called the eigenvector (complex) of matrix A if Ax = AX, where).. is co~plex

scalar and called Eivenvalue (complex) of matrix A corresponding to eigenvector x.

£xaD'lple 1: If A=[~ -~]. Find eigenvalve and corresponding eigerlvector.

Solution:·

Given, A = ll
ro -1]
O .
The characteristic equation is,
I_A- ll I = O

⇒ \-~ :~\ =O
⇒ )..2 +1 =O )

⇒ )..=±i (here eigenvalues are complex)

For)..= i,
Ax=u, x*O 'J

f
i.e. (A - ll)x = O , ,· ·,. · ·· alue )..,
.
havingnon-
trt·vial solution, then x is eigenvector, pf Eigv~v
•
-i -_1.) [xx1l = O.
or, ( 1 1 J , ,,
,. _ . :~, ,:, ...·, ·- c.~ P
.- .. . .. ·······: (i) .
- = 0
-ix1 - X2 - · ·
- . .. rt•t t.i.? ,:, •• •
.s~
1
• • ~
•
••••••••
(u··)
⇒ X1-ix2 =0
Here both equa ti.0 n (i) and
(ii) are identical.

Take equation (ii,, ••• I - ,-

%1 = ix2
Put %2 = 1 then Xt = i. • I • • r

1
. . 1
-[Xt = ( ) corresponding ).. = 1•
I• •
Hence, eigenvector lS x - xJ 1 r

For 1'. = -i,

(A-U)x=O

_O!) • G-m:J- 0
·' .::'.:.'.:.. (~ )
J

:::> ixt - X2 = ......... (iv) . ,

· -- O
xi + 1X2 tn·v1·al solution.
and 'denti'cal, sot·t has non-
Here, both (iii) and (iv) are t .. . ..

Taking (iv),
Xt + ix2 = 0
:::> Xt = -ix2 , '
(".
Put X2 = 1, then Xt = -i nding to X=--i.
x~
1 _
Therefore, x = [ x - 1
[•i] .15e1·genvector rcorresPo
!
"
<:
 MATHEIIATICS-D

0.5 -0.6] . . al f A and a basis for each eigenspace.

Example 2: Given A = [ _75
0 l.l . Find e1genv ue O _. _ .

Solution: •.l"f•_,.~_,.,. ,·. ·

·Let, .
A -[ 0.5 -0.6]
- 0.75 1.1
The characteristic equation of A is,
IA-A.II =O.
0.5 - l -0.6 ]
⇒ [ 0.75 1.1 - l = O.
• k •

⇒ (0.5 - l) (1.1 - l) - (-0.6) (0.75) = 0

⇒ l 2 - 1.6l + 1 = 0
1.6 + ✓(-1.6)2 - 4 X 1 X 1
⇒ l= 2
= 0,8±i0.6 . .. .. I • • •

For A = 0.8 + (0.6)i

(A-Al)x = 0

I ⇒

⇒
⇒ (
-0.3 - (0.6)i
0,75
-0 6 \ ( Xt\
0,3 - (0,6)i) ~J = ,0 • ;
(0.3 + (0.6)i)Xt + 0.6 X2 = 0
(0.75)X1 + (0.3 - (0.6)i) X2 = 0 ......... (2)
I

....... (1)

Here, both (1) and (2) are identical so it has non-trivial solution.
Taking (2), § / • ft I . 1, I

(0.75)Xt + (0.3 - 0.6i)X2 = 0

⇒ (0.75)Xt = -(0.3 - (0.6)i)X2
-1 I •

⇒ Xt = _75 (0.3 - (0.6)i) X2

. .},.

Put, X2 = 5, then Xt = -2 + 4i. . · '; J

2 4
Hence, eigenvector x = [;~ = [- ; i] corresponding to eigenvalu~ 1.:= 0.8 ~ (0.6)i.
And, the basis for the corresponding to 1 = 0.8 + (0.6)i is,

Vt {2; 4il "' '. ' : . ~ r' ~ ' . I •

For ~ = 0.8 - (0.6)i, (we can inspect directly V2 ~ 4i] conjugate of.v, which is ~lion
[-\ in
remark as below), · ·
(A-ll) x = 0.
-0.3 + (0.6)i -0 6 ·' \ [Xt7
⇒ (
0.75 0.3 +·0.6;j xJ =0 ,
'
⇒ (-0.3 + (0.6)i)x1 - (0.6)x2 ~ 0
,.;
....... (3) ·
 Eigenvalue and Eige
And, O:~Sx,1 + (0.3 + (0.6)i)x2 = 0 bector W e#fr-jJ
Here, both (3) and (4) . : · ; ·.······<4)
are identical so . h
it as non-trivial sol ti
Taking (4), u on.

0.75x1 + (0.3 + (0· 6).)

1 X2 = 0

=> 0.75 X1 = -(0.3 + (0.6)i)X2

-1
=> Xt = OJ5 (0.3 + (0.6)i)X2

= (-~-~i)x2

Put x':. = 5, then x1 = -2 _ 4i.

' . .
Hence, e1•genvector x =. [x~
x
= [-2 -4i]
5
· ~ -' .
corresponding to eigeny"1-'!e* ~ Q.8- (0.6)i.
- -- - I " :.. ~

2
A basis for the corresponding to A. =0.8 - (0.6)i is v 2 =[- ; ~] . ,I. f

Remark: From above example 27, it is noted that if A is 2x2 .ma~ of real entries, tlfen its ooniplex
-. ... eigenvalues (if exist) occurs in conjugafe pairs. And, '.tl)~ir f o~jr~nding eigenv~ors ~

ec :.
.....
,,....., .are also con1ugate.
'-- . So, m >· c· •· -eigenvector __, x =
. exampIe 27, after . rinding.
5
[-2 -f 4i]
corresponding eigenvalue = .0.8 -+ (0.6)i; we can inspect directly the eigenvector x =
[~-~]
5
for eigenvalue A.
..
= 0.8 - (0.6)i, because 0.8 ~ (0:6)i!is conjligate:fot, 0.83 + (0.6)i.
I'

Real and Imaginary parts of vectors

' ' • • I-
The complex conjugate of a complex vector X in en is the vecfor i ~in' en whose entries-are the complex
conjugates of the entries in x. The real and imaginary parts ~f a ~omplex v~tor. x are ~e vectors ReX
and Imx formed from the real and imaginary parts of the entries o~ x. - -

Example 3: If x =
2-i]
!
[3 4 i =
[2]
~ +
[-1]
!~ then, I '

A= PCP-1,

where P -= [Rev Imv] and C = b [a -ab]. I

I
!
 IIATIIBIIATIC8-ll

Example 4! Find an invertible matrix p and matrix C of the form [:

-bJ
a such that given tnatruc

A = G;~ has the form A = pep-1.

Solution: Given that

A=G -;)
The characteristics equation is,

IA-UI =O

⇒
1-A
1 1 3-A -O
-21-
⇒ -42 - 4A + 5 = 0
,L
4 + ✓(-4)2 - 4 X1 X 5
So, A= 2xt
• I
'1 ,.
⇒ )..=2,±i.
e- =-o-.
So, let).. =2 - i; hence a =2 and b =1. . ··-c. ~
-=,r r1;
'1 . • ) ' .'

' Thus,

r,j ' •• r: . .

For the eigenvector v and fQr A =!: 2 -, i, . ' ... ' :, •. 1 .J --

(A-AI)x = 0
-1 + i -2 \ (xi'\
:::) > ( 1 . 1 +;) ~J ..o.
f

! J , '
=> (-1 + i)Xt - 2x2 =:= 0 , , ..:.... (1)
=> Xt + (1 + i)X2 = 0 ...... (2)
Here, both (1) and (2) are identical, so it has non-trivial sol~tio~.-
\

Taking (2), 7

Put X2 =-1, SO, Xt =1 + i.

Hence,

V =Gj =[~;i] =LI:~]=[_;]+[~] i.

Therefore, Rev =[_;]and Imv = [~] .
 Thus,

P=[Rev - Imv]==[_~ ~]

yve have.already ~din above

C
~[21 -lJ
2 ·

, Here,-A = PCP-1, because -·

·-- Af'=n. ,;Jn ~J=L~ n

And PC= -1
[1 01][21 -lJ [3 l] .
2 == -2
1

EXERCISE 7 .5 ... ·
' eigenv~for of '"
Find the eigen value and c~rresponding · . .
f o·ven matrix A, which act on C2
(i) A
1 -2'\
= (1 3)
.. (1
(n) -2
5'\
3)
... ( 0 1,
(w) -8 4) ~
,
2. Find the eigenvalue and a basis for each eigenspace in C2 of matrix A.
. r r

.. G1.s2 -o.7"\ · · ·-~ - (i .10 8,

(~> o.56 o.4). -:~; ,~, <i.i >_. 4; -:2:v I I

3. · Find an invertible -matrix P and a inatrlx:C] ·of: the 1f6rm [~, · -:] stich that given matrix "A has the
form~= _pcp..i_ •"(', :it, p, I r !la . I • ,:

(i) [ 15 -SJ
1
.(ii) [51 -2J
3
, Fl "..-0.8]
(iii) L4 -2.2
ib ). · '[1.52 -0.7l
(iv) o.56 J o.4
' I., ' ; •. ., t=; )

l+i7 · , [l•lf-3i]•
1. (i) A = 2 ± i, [ _1 j · (ii) ii. = 2 ± 3i, 2

2.
1 1
(i) A = 4 + i, [ ; i]; A = 4 - i, [ ; i] .
. (ii) A= 0.96 + (0.28)i, [2; "] A= 0.96 - 28i, [2-i]
i ; 2 'J. ,,

[2-i]
(iii) A= -0.6 + 0.8i, [2; i]; A= -0.6 - 0.8i, 5 [1 -1} [4 -1]
3 2 -11 [3 -lJ (ii) P= 1 o c,"' , L4
- (i) P = [ 1 o , C = 1 3 [ 2 -1] -[0.96 -0.28]
.. . {2 -11 [-0.6 -0.81 . (iv) p = 2_ O· ., C - 0.28 0.96
(w) p 5 0 , C= 0.8 -0.6J
 7.8 DISOIETE DYNAMI~AL SYSTEMS

Eigenvalues and eigenvectors provides the key to. understanding the long-term behavior, or
evaluation of a dynamical system described. by a difference equation Xk + 1 = Ax1c, where the vector xk
records the state of the system at time k and A is a square matrix. The long-term behavior of the
these system is related to the eigenvalues and eigenvectors of matrix A.
Assume A is nxn diagonalizable matrix with eigenvectors Vt, v2, · ..., Vn and corresponding
eigenvalues A.1, A.2, . . . , A.n. For convenience, ass~e eigenvecto~ are arranged so that
IAt I ~ I"-2 I ~ IA.n I -Since (v1, v2, ..., vn} is set of eigenvectors so is basis for Rn, so any initial vector,

Xo = CtVt + C2V2 + ... + CnVn

n •.
By difference equation
Xt = Axo = CtAVt + c2Av2 + ... + CnAVn
= CtA.tVt + C2A.2V2 + ... + CnA.nVn .
Again,

X2 =Axt= A(CtA.tVt + C2A.2V2 + ... + CnA.nVn)

'.1L:..., ... ,_ ·;c ◄ .,4 v-;r~;.. :.c....,r;t,r, r ·"-, · J,, ~,"r·
1

= CtAtAVt + C2A2Av2 + ... + cnA.nAvn

( /

= C1At2V1 + C2A.rV'). + .-.. ,+ftrtA.n2Vn \ F

So, in general • . . i; I .,,. .., -.
I •

Xk = Ct(At)kv1 + c2(A.~)~v2 t ... + Cn(A.n)k Vn,.- r k = 0, 1, 2, ... ·

1
Let we see an example (Predator Prey ·system), in long-term behavior i.e. k~, what''x1t is? · :

Example 1: Denote the:_owl and wooa rat pop.u,latioJl#i.at.~e.-~k J,y ~ -=,~ , where k is time in
months, Oat is number of owls and Rk is number of rats (measured J!l JbousancJs) and

A = [ o.s
-p _l.lJ
P-4 be. predictor-~rey ~~
.., · of these two =-1
· poplUation.
1
(i) : Show that if the pre~dion ear;nneter p -~
1
o.i~ both population ~~w.
(ii) Estilpate the long-term growth rate of both populatic,n?
(iii) What is the ratio of population ~ owl and rat?
, I I '
Solution:

0.5
When p = 0.104, then A= [ -0.104
0.4]
1.1 · r-

Characteristic equation is,

IA-All =O

⇒
I 0.5.-A
-0.104
0.4I
1.l~A =O
(' .

'I
⇒ A2 - 1.6A + 0.5916 ~ 0
 This gives,

J' ; ;, .
. . . = 1.02, 0.58
Now, for A =·1.02, ' 1
rj !'

So, augmented ma--=~ .

u.J.A IS,

[A - A.I OJ = [A - 1.021 OJ
=[-0.52 0.4 OJ
l I
-0.104 0.08 o ·
~ [ -0052 004 ~] .
This gives,

and - (0.52)x1 + (0.4)x2 = o

_,,, 0.40 ,t - r. ;,
.... ,... __ ⇒ x1 =o.52x2 • f

\\i\1' .:a
J.O · .r, ... ,.. r·:::.
:....
⇒ X1 =13X2 1 ,.. -1 ,
.. · • . ., -·
~
- • •• _..J.

. • ~ .. J.
Therefore, x = [x1xJl =.[10/13 x:fl = ~~'[10/13~ ·A
X2 ;J 1 j• · •
, •i
I
1

. ; . I [10] _(f!'u(·.J L
Thus, eigenvecto~ corres_pondin~X.= 1.02'is•v_1 ~ 13 .
-► ~1•,•• '!'
...
1
! - ,_,., , -_•~ t, 1 ,::..
f ! ~ I • f • .";..' .a 1 !t • ' , .,J

I -. Similarly, fur l. =_iJ.58,'~ genveclor_is v;,= [:t ·:'.',' ;'

An initial Xo = C1V1 + C2V2 . I.

So, Xk = c1(A1)kv1 + c2(A2)k v2 ,· -1- ,\ • ' . ,)

\

,= Ct(l,02)k Vt + C2 (0.58)~V2 , .: '

=> :n = Ct (1.02)• G~] + c, (0.58){~]. . .,

j '

r
As k ➔ oo, (0.58)k -+O and for any ci > O;
.
Xk = Ct (1.02)k [!~] . . . ..(l) ...· '
I I,

I.

and Xk+t = Ct(l.02)k+t mJ • r I j

J,

. '.
~ (1.02) Ct (1.02i•T
!~T:, . I,

• 1.02Xk
 '
MATBEMATICS-D

i.e. Xk+l = 1.02 Xk ......... (2)

(i) From (2); both entries of X1t i.e. Oit and Rk grow bi- factor of 1.02 each month. Hence both
population of owls and rats are grow.
(ii) Also from (2), growth rate is 2% monthly. 1 - 0

(iii) From (1); when 0it = 10 then Rk = 13 i.e. for every 10 owls there are 13 thousand rats.
I

Applications to Differential Equations _

This section describes continuous analogues of the difference equations., In many applied problems,
several quantities are varying continuously in time, and they art: related by a 5rstem of differential
equations.
I'
~ = an X1 + ..... + a1n Xn
~ = a21 X1 + ..... + a2n Xn

I ~ " t f
....., - I

~
I
= an1 Xt + ..... + ann Xn • ')'/

Here xi, x2, ...., Xn are differentiable functions of t, wi~ derivatives ~,-~; ..· .• .'.., x:i, and the aij are
constants. ·- i j · i (" . , "

Now, write the system as a matrix differential equations.

. ..
x' = Ax .... (i)
• I
an ....... a1n
X1(t) X1 1(t)
a21 ••••••• a2n

where, x(t) = , x'(t) = 't • JI

Xn(t)· Xn'(t) ,, . ; I ant••••••• ann

• • • - ~ • ti • • i'• ' : ' • \ • • .J , I

A solution of (i) is a vector-valued function that satisfies (i) for all t in some interval of real numbers t ~ O.
- equation (i) is linear because both differentiaf of fun~on and m~tiplica~o~ o_f,. ecto1:5 by a
matrix are linear transfermation matrices. Since, if u and v are solutions of x' = Ax. Then cu +·
dv is also a solution, ~ause
cu + dv)' = cu' + dv' = c Au + d Au = A (cu + dv)

Example 1: The circuit can be described by the differential equation '

- c'..
.
v~(t)]
[ v~(t)
=[-(~
_L
1 + ~;)/Ct
R~J[vt(t)] ,
-2-- v2(t)
R2C2 R2C
where, Vt(t) and v2(t) are the voltages across the two capaci~ors at time t. Suppose resistor Rt is 1
ohm, R2 is 2 ohms, capacitor Ct is 1 farad, and C2 is 0.5 farad, and suppose there is an initial carge of
5 volts on capacitor Ct and 4 volts on capacitor C2. Find formulas for v1(t) and y2(t) that describe how
the voltages change over time. ··
 ,oJ11tiOtl• [-- (;, ) ..' ; C,
Given, A= ~I$J
R2C2
= [-1.51 0.5]
-1
x = [ ;~, and Xo = [!] -
For eigen values; solvin I A ,.
= - 0.5 and A.2 = -2. g
:i.1 - Al I = 0 we get

For lt..= . .P:5 we,get Yi= ~17

~2_j 1. -;. : I I4 ,
• I !fl

For A.~= - 2 we get V2 = ~1] [

' Jr(, '

. .e1gen functi ons Xt(t) = v eA1t

The · - '- . . ,
combination of X1 and x2. Set i and x2(t) = v2 eA-Jt both satisfy x' = Ax, and .so does
. any linear
x(t) = CtV1 eA-tt + C2V2 'eA-it = Ct [1]
2 -0.5 t [-lJ
m.
e . ii c2 e-2t

Since, Xo =m i.e., x(O) = So . 1

1
[~]=Ct[~]+ c 2[ ~ ]
or Ct-c2=S
or 2c1 + c2 =4
Solving we get Ct = 3, c2 = -2
.J

Ul f
I 1

EXl~llCISI~ 7 .8 I
J . -

1. .Denotes the owl and wood rat population at time k by ~

Xk _ (~ , w~ere k is m~~th, Cit is
4
number of owls and Rk is number of ra~ (measured in thousands) and A= -p o.
1.1 [O.S ]be
predator-prey matrix of these two population.
(i) ff p = 0.2, does the owl population grow or decline?
(ii) What about wood rat population.
2. In old-growth forests of Douglas fir, the spotted wol dines mainly on flying squirrels. Suppose
if . . [0.4 0.3]
the predator prey matrix for these two populations 1s A = -p 1 _2 . Show that the pr tion
eda . ·
 param~ p is 0.325, both the population grow. Estimate the long-term growth rate an'.d
eventual ratio of owls to flying squirrels.
I
3. The matrix A below has eigenvalues 1, 2/3 and 1/3, with corresponding eigenvectors v 1, v~ v3•

i.e., A=½ [~ t !]. =[tl v, V2 =ml =[~J

v,

Find the general solution ~f the equation Xk+t =Axk if Xo = [A].

-2· . . ..
What happens to the sequence {xk) ask ➔ oo?
4. Suppose a particle is moving in a planar force field and its position vector x satisfies x' = Ax

where A s [ ~ -:J, Xo = [~::] . ,

Solve this initial.value ptoble~ fort )~ 0. r, u 11
f ,,

1. (i) Xk = C1(0.9)k [!] + c2(0.7)k [f] as k➔oo,

(0.9)k and (0.7)k both tends to zero,
. : ) J=

XJc➔O; so both owl and wood

rat population decline. ,.
(ii) Wood rat population perish also decline.
Eigen values are 1.05 and 0.55.
-
•,
2
(i) (0.55)k ➔ 0 as k➔oo, and 1.05 > 1, so.both population will"grow at 5% per year.

Since Xk = c1 (1.0S)k [ 1;], as k➔oo ,..

I I r

so when OJt = 6 then &t = 13 i.e. for every 6 spotted owl there are 13 thousand flying squirrels.

.3. x,=f:)+@)"G]+3(½)"[lJ x, ➔ 1l [ .·- - -~-

e~J
~
4.
~~+ ~~
-6 188
-~+-e-t
0 70

I1
□□□

f '
 ORTHOGONALITY AND I

LEAS_T SQUARES
'-. ,i •

•I '

r;•

lEAINING ·OurcoMES
I I '.
... ' ' )
.,.

After studying tbfs chapter you should be able to:

& Inner Product, Length and Orthogonality ·
& Orthogonal Sets
f
& Orthogonal Pro;ections
& The Gram Schmidt Process I I

& Least Squares Problems r • :- ..

~
Applications to Linear Models ,,.... . - .,,..
&. Inner Product Space ; .' ,
r ~,,
 IIATIIBIIATIC8-II

A linear system of equations Ax = b that arises from experimental data frequently has no solution.
Often an acceptable substitute for a solution is a vector ~ that makes the diS tance between A~ and b
as small as possible. liere we begin the study with distance and then sum of squares and then least
squares solution of Ax = b. Also, the concept of orthogonality and orthonormality are used to find ~.
Finally Inner Product and its application on specific area is discussed.

8.1 INNER PRODIJcr. :LENGTH AND ORTBOGONAU'ff

Geomehic concepts of length distance and perpendicularity are well known for 9V and 9V, are
defined here for !Jtn. These notations are defined in terms of inner product of two vectors.

Inner Product
ff u and v are nxl mahices then uTv be txl matrix where uT be transpose of u, which is called the

• - [11]
112 •
=
[VJ
V2 . _ .,·
' , J
.
•
inner product ·o f u and v: ff u = ~ and v ~ then the ~er product of u ~d vis

r- El8ll
5olu

This means uTv is same as the dot product between two vectors u and v.

The following definition concise this concept:

E
Example 1: Compute u.v and v.u when u =(2, 4, 5) and v =(-1, 3, -1). s
Solution:
Here,
u = (2, 4, 5) and v = (-1, 3, -1).
Then
u•v = (2, 4, 5) • (-1, 3, -1) =(2) (-1) + (4) (3) + (5) (-1) ~ -2 + ~2- 5 =5~
and v•u= (-1, 3, -1) • (2, 4, 5) =-2 + 12 - 5 =5.
Thus, u•v = 5 and v•u =5.

,B 1n general, u•v = v•u for any vectors u and v.

,
 ~ e s of Inner Product Ortbogc,llllllty llllcl Leaat -&q--

Vecfors
u
),w u•w + v•w .
)•v • c(u•vj == u•(cv)
_Oand u U•Oifanc1on1
Yif
In a00ve·properties (" )
thi iv we observ d th
We take . s property as length of th e e dot product of a vect, .
. e vector. or to itself, is non-negative.
of
led the

• this defmition implies llvll2 = v•v.

Example 2: Find the length of a vector (4, S, 6).
Solution: ·
Let,
V = (4, 5, ~). '
Then the length of v is
'! llvll = ✓,.._42_+_5_2 +-@
_ = ✓16 + 25 + 36 = -{'77 .

Normally, if we divide a non-zero vector by its length then we get a new vector called unit
vector and its direction is same as to given vector. Mathematically, if v be a vector in Rn then its

V
11v11 ·
Example 3: Find the unit vecto; along the vector v ·= (-2~ 1;0) and verify it.
Solution:
Let
V = (-2, 1, 0)
Then, ---.---- - - - _~
lfvll = ✓<-2)2 + (l)2 +(0)2 = ✓4 + 1 + o="s.
Therefore, the unit vector of v is, ..

.~J2~Jr =(/s·Js~o).
Verification:
Let
 IIATBBIIATICS-D

Now, the length of u is, ------=--- ~

~
4+1 -~
llull = . ({s)' +(:Js)' +(0)2 =\j!+s+ 0~ 5 =vl=l
V
Thus, llvll =
(-2-:VS' ....[s'1 0)~ f
be unit vector along the vector o v • .
. vector of ·v, sometimes such processes
In above example, we create u &om v as a umt 1S called
normalizing v.

Distance in !Jln
ff a and b are two real numbers then the distance in the real line
between a and bis Ia - b I- If u ·and\ , are in !/l,2 then u - vis also in 9l2.
Let B = u and c =v then from the parallelogram ABCD, the·length of BC
is same as AD. That is u - v is the l~ngth from A to D. This shows the A
distance from u - V to zero is the 'distance from u . and V. So, in 9l2
distance from u to v = I u - v I - ,, D
u-v
This definition in 9l and in !/l,2 has a direct analogue in R•.

Example 4: Hu =(2, 3) and v =(3, -1) then find the distance ~etween them.
' ,
Solution:·
· Given,
u = (2, 3) and v .= (3, 4). . .
Then
U - V = (2, 3) - (3, -1) = (-1, 4).
Now, distance between u and vis,
dis(u, v) = llu. - vii= ✓(-~) 2 + (~) 2 ~ ✓1 + 16 = -{17_. I • . .
Example 5: Find the distance between u =[ 1] and z =[1] • I

• I
I,. I • J•

Solution:
Let

u =[-}] and

Then,
 So, Ortbogc,Dality and Leut ---- .
-,-.ttMe#rija~
(u - z)•(u - z) = [!] .[!]-
-6
1

. -6 - 16 + 16 + 36 = 72.
Now the distance between
u and z is
dis(u, z) = llu II V
- z =cu ~ z)•(u - z) =fn. =6'./2.
Orth98(!nal Vectors
. .
Here we discuss or the fact th .
.
at the.concept of DP1-nAftdi
. r-&r-,• C
ular linem
. Rn.

--
>c
EllllD~le 6~ Show ~ t the vectors u -=(2, -3, 3) ·and v = (1Z 3, :..s) are orthogonal.
Solution: • . . - . , ...
Given,

u =, c2, -3, 3) and v = c12, 3/-s>

Now, ·

:= (2, -i 3)•(12, 3, -~) ~ 24 ~ 9-}5 •= CJ.

1

U•V ..
means u and v are orthogonal.

• Observe that the-zero vector is orthogonalrt:o,eadt-ivector u in.R because O•u = O.·

Proof: First suppose that u and v are orthogonal. Therefore,

u•v = 0 .... (i)

Since Huff= u•u. So,

lfu + vu2 = (u + v) ·•(u + ~) ', .
= u•(u + v) + v•(u + v)
= u•u + u•v + v•u + v•v

o o
= lfulr + + + llvll2 [using (i)J
2
= lfulf + llvll
Conversely, suppose that -
2
llu + vll = llulr + llvlr .J

⇒ (u + v) • (u + v) = 11u1r + Jlvl: 2

⇒ u•u + u•v + v•u + v•v = llull + llvll

 IIATBBIIATIC&-11

⇒ u•v+v•u =O
⇒ 2(u•v) = 0
⇒ u•v=O.
This means the vectors u and v are orthogonal.
Orthogonal Complements
We have already discussed about orthogonal. If a vector y is orthogonal to every vector of a
subspace W of Rn. Then y is orthogonal tow. We can search there may so many vectors that are
orthogonal to W, then the set of all such y, is called orthogonal complements of W.
ltlon (Ollltogonal Complements)
set of all orthogonal vectors to a subspace
W. It is denoted by wi and rea~.~ 'W peq,etl ·

- .

{ii) WJ. is a subspace ~f Rn.

Theorem 3: Let A be an m,
, null .Sp ..
· > · ··
-
{i) A vector xis in WJ. if and only if xis orthogonal to every vector in set that spans W.

'
null-spa
Exam]
Angles in R2 (or R3) Solutj

If u and v are two ¥ectors in R 2 {or R 3) and 1if 8 ·be;the angle between these vectors, then the _d~t. L
product of u and v is defined as, . \

u•v = Hull llvll cos8. , ,-

. - _, ~, , 0 - -
l
'
u•v .:.. _....,.,_
.....
⇒ cosa = Hull llvll '"'

⇒ 9 = cos-• CMiiMJ .
d .
Thus, the angle 8 between any two vectors u and v is defined as;
U•V l ' '
8 = cos-i ( Hull llvlV ·
Verification:
 Let u and v are the points . 9(1. OrtbogonaUty and Least 8quana W 1Caaffla81 .
figure. By the cosine law • m · Then llull, llvll and llu - vii are length of s1"des of triangle,
. -see
- in

1111
. - vlf = llulf + llvlf - 211u1111v11 cos~ (")
Smce u, v e !Jl2, 80 11 = ( ) .. . 1
' 111, 112 and v = (v1 2) Th 1.
· • ' v • en '( ) can be rearranged as
. 1 .
llull llvll cose = 2 (llulf + llvlf - llu - vlf) .

1
= 2[u•u + v•v - (u - v)•(u - v)]
1
= 2 [(u1, u2)•(u1, u2) + (v1, v2)•(v1, v2) - [(u1, v2) - (v1, v2))•(u1, u2) - (v1, v2)]
1 -
:z 2 [u12+1u.i- + V12+ vi- - (u1 - u2)2- (u2 - v2)2]
I
t

1 .
=2 [1112 + u:z2 + V12+ vi- , u12- v12~ 2u1v1- ui2 - vz?- + 2u2v2]

., l I

=u•v
Eumple 7: Find the angle between the vectors (0, -5, 2) and (-4, .;.1, 8).
Solution: •.., · '
Let
u = (0, -5, 2) and v = (-4, -1! 8) 8P?- e·be the angle ~tween u and v.
Here, ' '

u•v = (0, -5, 2)•(-4, ::·1, ~) = 0 + 5 +116 = 21

And, --- I'• --- _ro:;

. .llull = ✓o + 25 + 4 = ~ and llvll = ✓16 + i + 64 = "81 = 9 .

Now, angle between u and v is, ., I 7: i

s.

8 = cos-' C~il~1U = cos-' (9~) = ~os-' (~}

• v of u ~ (1, - 2, 2, 3) in the direction u.
Example 8: Find the unit vector . · , , , I..
' ,
Solution:

= i)
u --
V = I 1ul I
(1, - 2,2, 3)
= -{! _+ 4 +_4 + 9
{1, - 2, 2, I 31
=3'/2
 IIATBBIIATIC&-D · . ,

1_BXERCISE 8.1
I •
1. Using the vectors, compute the quantities where

U =[i], =[!],
V W =[;!], =[~j]X

1
(i) v.u (ii)
u.u, v.u, and~ w.w, x.w, and ~ x.w (111···) -w.w w
1 (iv) - u
u .u

u.v~ v
(v) ( v.v) ·
. (vt") (x.w)
x.x x (vii) llwll (viii) llxll
2 Find a unit vector in the direction of the given vector.

(i) [ !] (ii) [ !] (iii) [i~~] (iv) [ 8/3]

2

3. Find the ~~ce between x =[~~]and y = [ =;]

4. Determine which pairs of vectors are orthogonal.

(i)

5. Find the angle between given vector~:

(i) u = (1, - 3).and v = (2, 4).

(iii) x = (1, or 1, O) and y = (-~, -3, -~, -3).. ..

..
6. If u, v e Rn, prove that [dis(u,·-v))hJ [dis(u, v)]2 iff u.v = 0.

1. (i)
8
5, 8, 5 (ii)
1
35, 5, 7 (iii)
',.
.r/35l -1/35
-1/7
(iv)
[-1/5]
2/5
,1

(v) [ 8/13]
12/13
(vi)
.[-10/49
30/49l 1
(vii)~-
.. (viii) 7 . . 't . '.J,"':

15/49 I
-6 7
v6i ~
4· .
[U
2 (.

2. (i) [-3/5] (ii) (iii) (iv)

4/5· v6i ~ 1•
'
-3 4
v6i ~
s{s
3.
(iv) No
4.
5.
(i)
(i)
No
31t
-4
(ii)
(ii)
Yes
' c2or
cos- 9'/10
1
(iii) Yes
(iii) -31t4·
 Orthogonality and Leut a - . ~ -s--WIC&Amaat
,., 011TBOGONM.. SETS •
~
We }lave already discussed that two vectors u and Orth 0
e,ctend the concept toward a set of vectors {u u2, v are
u } hich ogonal to each.other if u. v = . If we
1
nltlon (Orthogona Sei) ' · · ., P ' w we called orthogonal set.

p \ult :U2,
•.,~
of • Up} in tl• is;said to-be_ an Ollhogonal oet If 1Jt'Ui s ofor i ,. I 1

u .. = (-6 3a 9)
Example. 1: Examine d vectors {ui, U:z, '13} is
set of . an orthogonal set where u1 = (2 -7 -1)
• • - • an u, = (3, 1,-1)? , • ,

Solution: Given
. Ut = (2, -7, -1), U2 = (-6, -3, 9), \13 = (3, 1, -1).

Now,
U1;U2 = (2, -7, -1~•(-6, - 3, -9) = '12 +·2t.: 9 = 0.

u2•ll3 = (-6, -3, 9)·(3,

,- . - 3~.. - .9 .= -30 ,t;: 0.
1, -1) = -:18 ~
• o • • I

Ut•\13 = (2, -7, -1)•(3, 1, -1) =6 - 7 + 11 =0.

This shows th.at the set {ui, u2, \13} is not an orthogonal set.

- The sets {u1, u2} and {ui, -\13} are orthogonal.

n then S is Ilneatly

Proof: Let S ~ (u1, ... , up} is an orthogonal set of non-zero vectors in R• Then .
Uj = o· for i * j and . i,·j -~ 1,
1

Ui
0
: : ., 'p ..
Choose the scalars c1, ... , cp such that
CtUt + ... + CpUp = 0. ;.J'•

Then,
(ctUt + ... + cpup)•Ut =O•ut =0
⇒ Ct Ut •Ut + C2 u2•Ut + ... + CpUp•Ut =
0

⇒ Ct Ut •Ut + C2 Q + · ·· + Cp Q = O , .. I ••

⇒ Ct Ut•Ut = Q.

Since u "O. So, u,•u1 "0. 'lherefore, c, = o.

1
!_:, .....1 all are zero •
Smwc:u y, C2, ••• , Cp nd
Thus, the set S is linearly indepe ent.
 IIATBBIIATICS-IL

Example 2: Show that ((3, 1, 1), (-1, lr 1), (-½, -2, i) is an orthogonal basis for R 3
•

Solution:
Let

Ut = (3, 1, 1), U2 = (-1, 2, 1), U3 = (-½, -2, ~) ·

Here, u

1 7) 3 7
Ut•\13 = (3, 1, 1)•( - 2, - 2, 2 =- 2- 2 + 2 = 0.

Therefore, {u1, u2, \13} is an orthogonal set.

Here {u1, U2, U3} is an orthogonal set of vectors, so {u1, u2, 113} is a bas~ for !IV and therefore, is an
orthogonal basis for !/l,3• ·

5: -=----=-=-=---:--1

Proof: Given {u1, ..., up} be an o~ogonal b~sis for a sµbsp~ce yv of Rn; T_h~n, for i ;tj,' ·
Ui•U; =0 for i, j = 1, ... , p' . . ',: '
Let ye W then y can be written as linear combµ1ation of {u1, ..,. ~! up}- That _is f~r some scalars c1,
••• , Cp such that · · ·
y = CtUt + ... + CpUp '' .
Then, . ,_

y-ut = (ctUt + ... + cpup)•ut

= Ct Ut •Ut + C2 U2•Ut + ... + Cp Up•Ut
= Ct Ut •Ut + C2 0 + ... + Cp 0

⇒ c1=
.I.:!!!.

Similarly, we can ob~

q= .Y:!!i. for j = 2, 3, ... , p

Uj9Uj
Thus,

q= ~ for j = 1, 2, . . . , p.
1;1tllj
 Orthogonality and Least a -.. __
An Orthogon~ Projectio~ ~--- w ICaAmR e1
I
Let u be a non-zero vector in Rn C . · ·
· ons1der as a bl
two vectors as · pro em of decomposing a vector Y m
• R n into
. sum of

y=au+z
. .. . (i)
for some scalar a., where z 18
. 0 rth z=y-y"
ogona1 to u. '
Since, z is orthogonal to u . So,

z•u=O ⇒ (y-au)•u = O
0 ----,,....__ _. u
⇒ au•u =y•u y=au

⇒ a= ~
u•u
Then. j •

. .. (ii)

That is,

"
y = Proju(y) - - (~J·
= ·· ·
u•u
u.

The vector y is called the orthogonal projecti~ of y

onto u and the vector z is called the component of y
orthogonal to u. · . · · ' · ··
- .

Example 3: · Let y =[~and u~ [!] .Find the ort~ogonal proj~on of y onto u.

Solution:
Here, ·
y-u = m-m = (7)(4) + (6)(2)
. '

=.28' + 12= 40
> \ •
-
I • '

m.[~J
) I S

and u.u = = (4)(4) + (2)(2) = 16 +4= 20.

y
Now, the orthogonal projection of y onto u is,

y=(Y:!!.Ju=:m =2 m=m.
u•u

■ Since the vector notation of line joining [-n (i.e. (-1, 3)) and origin is [-n ·-
 "IIATBEMATICS-D •

Example 4: Find the orthogonal projection of [ ~l] onto the line [ 1] and the origin. l f

I'
Solution:
Let,

Then

[-lJ [-lJ =
y•u = 1 • 3 (-1)(-1) + (1)(3) .
=1 ; _3 =4
-- - -- u•u =[ i] ·[i] =(-1)(-1) + (3)(3) =1 + 9 =10
Now, the orthogonal projection yof y onto u is,

Y-(!:) 0 -@[i]=¾[i]=[fj;J.
Example 5: Let y = (3, S, -1) and u = (4, 1, 3). Find the orthogonal projection of y onto u. Then
write y as a sum of two orthogonal vectors in which one is Span{u} and another is
orthogonal to u.
Solution:
Here,

{u} = y. Here,
28 7 21~ (11 58 -~, 308 406 714
y•(y- y) =( 13' 13' 13).. 13{13' _13 ), ~ ! 69 + 169 -169 =o.
- . } . - "'

This implies y is orthogonal to {y-y). Therefore, (y-. y} is orthogonal to u.

Thus, y is the sum of Span{u} at'!d orthqgonal vector to u.

that the standard basis {ei, ... , en} for R", is an orthonormal basis for R".
 Orthogonality and Leut &quan. W r = - -.... 1
Show that lv1, v2, v3) is an . ICaAPiiit el
3 orthonormal basis for R3 where _........,

Solution:
Vt• ( - _!_ l :'\
vii' -./ii' -./ii)' Y2
( 1
a ~•
l
i• ~),Vs= (i, i, j}
Here,

l I

· 9 1 1 11
lfvill = V1•V1 = 11 + 11·+ 11 = 11 = 1,

1 4 1 6
lfv2II = V2•V2 = 6 + 6 + 6 = 6 = 1,
1 16. ! 49 66 . )'10. }. . )
lfv3II = V3•~3 = 66 + 66 +-~ -= 66 = 1. ·,r! •.,
~ ' }( •' J, • ~ J .

Since we know every orthogonal of no.n-zero vectors fomr a basis for its v:ector space. So,
{vi, v2, v3} is an orthonormal set. · °' · , .~. ... .1 •1
, .
Example 7: Determine given sets of vectors are' ~rthdnormal? H a set is only orthogonal,
normalize the vectors to produce an ortti~normal ~i,-
f.

Solution:
Let,

1/3] [-1/2] .'

u = 1/3 and v = ~
[ f/3 1/2
Here,

u•v = [!~~] •[- t

1/3 1/2
1 2
]= (½) (-½) + G) (o) + (½) m=-: +? ~¼ =
O
;j

naI vectors Also ·~ ·· ·

This means {u, v} is set of orthogo . ~ . - .

1/3] [1/3]-!+!+!=~ ~!~·i

llull = u•u = [ 1/3 . 1/3 - 9 9 9 9 3
1/3 1/3
-1/2] [--1/2] 1 ' ! -2-r-
o --+o+ _! "'"l·
and llvll = v•v = [ 0 • -4 4 '~ "
1/2 1/2 .
 This means u and v are not orthonormal vectors.
Now, normalizing the vectors u and v as

u• = 1M1 :: _ r
"\JU•U
~ ¾[~~~] [~~1]
=_
-vl/3
- _
-v3 1/3
=
1/~

and v' =1;;1 = (~

1 2
f--"l v ¾[- 6] [-l~~]
=_ =
\ -Vv•v) -v 2 1/2 1/'V2
Then clearly {u', v'} is a set of orthonormal vectors.
-~---~~--=-::-:-:-::~ ~:;-:"""":--T,--:-""~'""'"77
&t a,.,cn tnatrbfll ~~it if U U • t
Proof: For simplicity, we prove the theorem for 3><3 matrix in which each column in Rm. Then, the
idea leads to complete proof of the theorem. Let U = [u1 U2 113].
Here,

UTU a[:~ [u1 u, u,]

u1Tu1 u1Tu2 u1Tll3]

= u2Tu1 u2Tu2 u2T113
[
113Tu1 113Tu2 113T113
Therefore, the columns of U are orthonormal if and only if
ul Uj = 0 for i -:I:- j for-i = 1~ 2, 3.
And, the column of u all has unit length if and only if
u?ui =1 'for i =1, 2;3. •- '
This means U has orthonormal columns if and only if 'I

uTu=[~ ~ .~]=1. .
0 0 1
I
I

1/{2
'J/3] .Then show that Uhas orthon~ columns if and only if UTU = L
-'J/3
LetU= [ 112 1/3
I

Solution:
Let

Where,
11-{2] [ 2/3 ]
u = [1/[2 and
1 -;1; · U2 s

Then,
11-{2 2/3]
-2/3
-2/3 1/3
 1 Clrthoco1181ity ..... 1 - ~ U. '2§§81
.-f: ' - ·[1/2-fll/2+0
- 2/3{2- 2/3'.[2 + o 2/3'{2- 2/3'[2 + OJ
r , 4/9 + 4/9 + 1/9
I
=[~ ~]
=I.
Next,

llt•Ut 1/f2 .[11,~

.=[i1{2] 1/f2j =½+½ + 0=l

·r [2/3] -(2/3]
• 4 ·4 1 ..
U2•U2 = -2/3 . -2/3 =-+-+-= 1
9 9 9
--
l , . , . -~ ,r. [1/{2] [2)3]
1/3 1/3
. J • •• • .

U1•U2 = 1/{2 . -2/3 = :~ - :~ + Q = Q

I • • ,-
o 1/3 3·v2 3-v2
and similarly, uru1 = u1•u2 = 0.
This shows that the columns u1 and u2 are orthonormal columns of U.

Example 9: • Verify that IIUx!I = llxll where·U is defin~d in'~bove ex~ple and x = [f].
Solution: .
"t. I ' ! .
From above example,
Here,

.Ux .
. [11/z
1/-{i
=.
2/3·]
-2( 3
[fi]
3
. 0 1/3

. ~ [~ ~ i]-=[!i] . ... I

··· 1 . , , 1
, • vr
· :
Then, . l ' ( ..
ltUxll =:J9 + 1. +1 =fii .. ,
.-
' .
'
And,
llxll = ✓z+ 9 =·{Ji
Thus, IIUxll r- -fii. lfxll -

..I_EXEllCJSE 8.2 .
 .IIATBBIIATIC&-11 .
2. Show that {u11 u 2} or {u11 u2, \13} is an orthogonal basis for R2 or R3, respectively. Then~ x
as a linear combination of the ~•s.

(i) Ut=[!J,u,-[:J,andx =[!1] (ill) Ut =Gl u, =[1], ~[t],andprn u,

(li) u,-m. u,•[7J,am1x=[t] (W) Ut =rnJ u,=[H ~=[!],and x=[f]

3. Compute the orthogonal projection of[~] onto the ~e through[ i] and the origin.
4. (i) Let y • [;] and u • [-~] . Write y as the sum of two ortho~onal vectors, one is Span {u} and
one orthogonal to u.

(ii) Let y -[~] and u = [~ . Write y as the sum of a vector in Span {u} and a vect~r orthogonal
tou.

5. (i) Let y =[~]and u = [!J. Compute the distance' &om y to the line through u and the origin.

(ii) Let y =[;]and u= [~] . Compute the distance from y to the line,. through
. 4..
u and the origin.
•
6. Determine which sets · of vectors are orthogonal. H ~ set is o.nly orthogonal, normalize the
vectors to produce an orthonormal set. , -· ,. . ,

mfl.[!~~]
~ --·

m.m
\

(i) (li) [ -o~s6],[~::J, (ill)

1
(iv)
1110
3/"2f> , [-t/"2fJ
3/VW] , [-1/'V2
[3/....{2o -1/V20 1/'V2 o ] (v) [~~~],[_ if'l,r}t].
11-{IB -1/~J t-2/3
•

1. (i) No (ii) Yes (iii) No (iv) Yes

(v) Yes <~) No

2.

3. [-12] . I • •• 1

4. (ii) [~;;] + [~~;]

5. (i) 1 (ii) 3-\/5

6. (i) No, (ii) Yes

(iii) Yes (iv) No (v) Orthnormal

 ProJection of .
a Straight line or a curve in R2 .
IS a straightline.

- ax is orthogonal to y-a.x.

n.

he

■ The vector yis called the ortho . . wntten as Projw..(y).

I.-----=-go_n~~~p~ro~Je:c:ti:on::_:of~y~o~n~to:W~an~d~~- .
- A .
Z -y-y y

The orthogonal protection of y onto W

lrample 1:
Let Ut = [l], rJu, = [ and y = m. Observe that {Ut, 112I is an O(lhogonal basis for

W = span (ui, u J. Write y as the sllDl of a vector in Wand a vector orthogonal tow.
2

loiutton:
The orthogonal projection of y onto W is
... (i)
1
y= (~lu1+(~ u2
u1•uJ u2•uJ
 MATHEMATICS-II

Here

Y-~• =m ·[I]= 2 + 10-3 9 =

y•u, =m-[-n = -2 +2 +3 = 3 I I

u,•u, =[l] ·[I]= 4 + 25 + 1 =30

Then (i) becomes,

Also

twnple 3

Since W = Span {u1, u2}. So, y is in \ v. Also, by OrthogonaI°Decomposition Theorem, y -yin

orthogonal to W. •
.. '
r ., I .

And,

y
-
=y + (y-y) =[-215]
1/5
~5].
+[ 7
14/5
~lution:
Hert
TheoY8m 9 fib• lest Approximation Theorem)
be a subspace of R", vectot in.R! y onto
loset po· · the
lly-ylJ

,.
 Ortbogonauty 1111d
2
f;jBlllPle 2:Let U1 =[;], =[~
U2 ] and -[1] 1-at &qa.._ 1JJ ~ ,.J/i/ §iari#!
- -1 1 y - 2 . Observe that u .
. . W :-· sp~ (ui, u2}. Find th I 3 { i, u2J is an orthogonal basis for
50Jutis~n. th I e c_osest point in W toy.
mce e c osest point . W

Here Y ~~:~: ➔ (~~}, •• (i)

y•u, =m ·[IJ= 2+10 -3 =9

2
y•u, =m-[~]=-2+ 2+ =3 3

u,•u, =[1~ HJ=~+ 25 +1 =30 f, t

U 2•U2
· =[-2] [-2] .
.1 • 1
1

~ 4 + J. + 1 = 6
I • r

,
1' I

1 _l ~-... .
Then (i) becomes,

Example 3:

,,. .
-ym
. .. ~

Write x as the sum of b\TO vect~rs, one in Span {ui, u2, 113} and the other in Span \114}.
f • l ~ •

r
Solution: I ; \

Here,

-10- -3·
-8 5
2
•
1 + 0) (11 30-40+2+0 _:!=-~
Q.O) (3) + (-8) (5) + (2) (1)+ {) (1)= 9 + 25 + 1 + 1 - 36 9
0 ,__l
-3- 3- = (3) (3) + (5) (5) + (1) (1) (
5 5
•
1 1
...l_ i-1_
 MATHEMATICS-II

~
ll3•ll3
[il[!J (10) (1) + (-8) (0) + (2) (1) + (0) (-4) · 10 + 0 + 2 + 0 =.12 = ~
(1) (1) + (0) (0) + (1) (1) + (-4) (-4) - 1 + 0 + 1 + 16 18 3

[i]-[i]
and - -
X•l4 [!J I~J (10) (5) + (-8) (-3) + (2) (-1) + (0) (1) _ 50 + 24 - 2 + 0
(5) (5) + (3) (3) + (-1) (-1) + (1) (1) - 25 + 9 + 1 + 1
3726 = 2
\4•\4

[i] {1]
Let x1 is spanned by {u11 u2, \13} and xz is spanned by {\4} then~
x1 + x2 = Span {u1; u2, \13} + Span {\4}

= [( x•u1_! u 1 + (x•u2_! ~~ +(x•113)] ll3 +(X•\4 ·\4I

lu1•uJ u2•uJ . 113•\IJ \4•\4 )

.(~~r~1J r:\-2~l!r;J._(~\[~'.J [:3J

= 9) ~ 9/ + 3) ·: ~
2
_1
1

=@[<~>[}]+c~2)m·+.(6)[i]~(18)[~]] l _J

' _ (!~ [-
- \CJ)
~ ~ ~;:; ~54] -_-(\9)~'\ [~
32 - 2 + 6 - 18
-~2]'
_[~~] _·,-
18 - 2 - X
8 - 2- 24 +_18 0 0
This shows xis a sum of x1 and x2 where x~ =Span {u1, u2, 113} and x2 =Span {\14}.
Example 4: Verify ttiat {ui, u2J is ~ orthogonal set, and t~en find the orthogonal projection of

y=[JJ ontoSpanfu~u,twh~u,·= [:J: ~-[!].

Solution:
Here,

U1•U2 =G] ·[~ =(3) (-4) +(4) (3) +(0) {O) =-12 +12.+O=O .
This means {u1, u 2} is an orthogonal set.
 And,

U1 •U1 == [!] •[!]-

0
3
,
O - ( ) (3) + (4) (4) + (0) (0) = 9 + 16 + 0 = 25

U2•U2 == [t] •[t]-

0 O - 16 + 9 + O = 25

y-u, =EJ·m=l8 +12+0=30

y-u, =[l]·[1]- -24 + 9+ o =-15

Now,.the orthogonal - . ,..

Pr0Jection y of y onto span {u1, u2} is

y -(~:}, + (!:::9 u,

. ,... .
(18/5) + (12/5)] I •

=[ (24/5) + (-;~/5). _ I -
0 +O '

Ji .: .I ' I
.
. . .... . .",
• 1. ,, .

Example 5: bt w~be the subspace-spa~ed by...the :u's, ana write y ai.the--s1UD-ota vector in W

and a ~ orth~;., ~
.
j.~;{:J ~, {} ~ =[
W
.
!l ~ ,
Solution:
Here,

.'
 IIAT.JIEIIATI~D

Then the orthogonal projection yof y onto Span (u1, u2} is , •I

y = (:.:0 Ut + (;::},

2 + (-1/2)]
=[ 2 + 3/2
2 + (-1)

and

Now

and

This shows y is a sum of two orthogonal vectors {y - y) and ywhe~e yis in W when W be the
subspace spanned by u1 and u2. G
Example 6: Let y = (-1, -5, 10), u1 s::: (S; -2, 1) andru2 = (1, 2l -1).-Find ,the nearest point in W toy
and distanc_e between y and'lthe hearest point where W = Span {ui, u 2).
Solution: r ,
i- - ! •·
Let ybe the nearest point in W = Span {u1, u2} toy. Then

= ((-1, -5, 10) • (5, -2, l)J ( _ ~) + ((~1, ·-5, 10) • (1, 2, ~1)\ .
5 2
l (5, -2, 1) • (5, -2,.1) ) 1 1
l 2
(1, 2, -1) • (1, 2, -1) ) (l, , -l)
(_-5 +·10 + lQ'\ (-1-10- lq"\
= \ 25 + 4 + 1 ) (5, - 2, l) + 1 + 4 + 1 ) (1, 2, -1)

=@,-1, ½) + (1, -7, 0

= (-1, -8, 4).

And the distance from y to the nearest point y in w is IIY _ A I' · b the Best Approxunation
Theorem. Y" Y
 Here,

Y-y"-(
- - l, -5, 10) _ .
Then ' . (-1, -8, 4f == (0, 3, 6).

llr-yll
Thus, the distance betwee
-vo
-
9
+ + 36 '4s
== -'I 'K> . ,

Example 7: Find the b n y and they is~ . . .

estappro . .
Xllnation to z by vectors of the fo

z-[t}v, =[1} v,e[t]-nn~v, +c,v,wh=

Solution:
Here, \

. ..,

.I .-
I' -

I I
..) J
'

_J .

Thus, the ~t ag t o x ~ ~ z by vectors of the form~= c'..v' + c, v2 1is (1.,o, -; '. 1)-
Eumple 8: tety=GJ,u, =[~] andW= Span(u,J. _,., . ,

(i) Lei U be t:lte 2X1 inatril< whose only column is u,. Compute uTu aitd ml.
1
(ii) Compute Projw (y). • J

Solution:
..
'
'
(i) Let : .I

Then,
 ·IIATBBIIATIC&-11

Now,

UTU--[-1 -±J[l/{io]
{10 {ii) _3/{ii) =_!_ 10 =1
10 +~

and T [ 1/{io] [-1 1/10 -3/lOJ

UU = -3/-{lo -{lo {ii) - -3/10 9/10 ·
_±] -[
(ii) Here,

Now,

Pro· (y) =(.I:!!!.

jw lu1•uJ Ut
=1 (-201:®)
1
11{w] = [-2]6
[-3/,{io
' .
and we have Projw (y) = (uuT) y.
So,
.- ._

I
I •

/1_EXERCISE 8.3
1.
. ·
Assume that {u1, ... , 114} IS an o~ogonal bas~s for R\ ~t u1 = ~
•I
:-- · -• [lj 2
u2 = -; l-21
1
'13_=
ll~ .Ll}
1
:~ '114 =
1
:
2

v =[ j]- Write v as the sum of two vectors, one in Span 1"11 and fue other in Span {u1, u,, U:,}.

• : ) • " ,. :, '- I •

2. Verify that {u1, u 2} is an orthogonal set, and then find the orthogonal projection of y onto Span
{u1, u2}. , . ., . •• !t·'"''

{i) y=[!Ju,~mu,=[!] . • I

•• •J

(iii) y-m -r~J =mu, u,

• 'l
 3. Let W be the subspace s Orthogonality and Leaat ~ W ICBAmasf
o~ogona} to W. Panned by the u's, and write y as the sum. of a vector in Wand a vector

· [1] [1] . -
~ ~ =[~}ti,= [_t} ~
· . · -· . , 5 . 4 1 -1 -1

' V(1) - U:3= :2 U, =[!] . • (ii) y =[ 1} Ut =[ !]

(iii)
y [}~ [~1} =[1}~ =[OJ-~
= 4 _. 1
u, - u,
1
l1J
-1

4. Find the closest po· t to .

m Y m the subspace W spanned by v1 and v2.

(i) y=m.v. =[Hv,=[1] (ii)_. y ~[Xl v, =rn- =[IJ v,

5. Find the best approximation to z by vectors of the form c1v1 + c2v2 when

.
~
-
... -.... ..._. - ~--'

n
C"C -

6. Let y =r:J u, := [!J u, =[ ~ind ~e ~~ from·; ·~th~ plane.in R' ~ e d ~ and U1

112-

7. Let y =[!l, =[ij;J~ =[;~~

1J
u1
2/3
u2
3

1/3
} and w = sP'!Il /ub u,1.
·
(i) Let u = [u1 u 2]. Compute uru and uur.

(ii) Compute Projw{y) and (Ull)y. .

• I •

3. (i) r1o/:J r-:1:J

8/3
+
7/3
(ill) [!1 {fl
4. (i)
_s. [11 · 6. ~
(ii) y= (2, 4, 5), (UUT)y = (2, 4, 5).
7.
 ◄
11&1+ enca a n.

The Gram:-Schinidt process is a simple process or algorithm to obtain an orthogonal or

orthonormal basis for any non-zero subspace of Rn.
Example 1: Let W = Span ix.., x2) where x1 = (3, 6, O) and x2 = (1, 2, 2). Construct an orthogonal
basisforW.
Solution:
Let Vt = Xt. Also, let

.... (i)

Here,
~ _ r1,·2, .2)- (3, 6,'o) _3 + 12'+ o 1
Vi·Vt - (3, 6, 0) • (3, ~, 0) - 9 + 36 + 0 3
Then,

V2 = (1, 2, 2) -½ (3,,6,_0) = (1, _2, 2) - (1, 2,'0) = (0, 0, 2).

Now, ·t he set {Vt, v2} is an orthogonal set for W. Since W ; Span{x1, x2}, so the dimension of W i.e.
dim(W) is 2. Therefore, the set of two vectors {v1, v2} is ~ orthogonal basis for W.

xi-x1J . (x2•v1J . . . . . , '

Note: In above example the f~rm (xi•xJ ~: can r~~~e .~~j~vi~vJ v/ whi~h ~ -~rpJec_
tio11" \

en {v~ •• -.
In ad,dition, ·
Span {v1, ..., Vk} = Span _xi, ..., x .
 Orth .
Example 2: Let
X1 == (1, -4, 0 1) a d
ogonauty r-----
lllld Leaat Squares w lCHAmR 81 ~2
}
orthog al ' n = (7 7
X2 - -
on basisforw ! - , -4, 1). If w -
Solution: by using the G S - Span
(x'II x2I then construct an
. ram- chmidt process.
G1ven x1 == (1 -4
R4 L t , , 0, 1) and X2 = (7 7
· e V1 == X1. By Gr ' - ' -4, 1). Also· let w - s
basis for W atn-Schmictt process w ' - pan {x1, x2}. Then Wis a subspace of
Tak · e construct vectors
· v2so that {V1, v2} IS
• an orthogonal
e . V -
.' . . i-xi==(l,-4,0,1)
and v 2 =: X2. -(X2 •V~'
Vt
Vt ~ V

-_ X2 - (X2 • X~
-----
Xt • X
X1
·_
[' .' Vt = X1]

=: X -((7, -7, -4, 1) • (1, -4, 0, 1))

2 ~
(1, -4, 0, 1) • (1, -4, 0, 1)
(Z+28+0+1)
= X2 - \J + 16 + 0 + 1 Xt .

= (7, -7, -4, 1) _ ({ID (1, -4, o, 1) .

= (5, 1, -4, -1). .
This, {v1, v2} is an orthogonal set of non-zero vect~rs in W. Since Wis defined by abasis of two
vectors. So, the set {vi, v2} is an orthogonal basis for W. ·
Orthonormal Bases
The following example shows a process to construct an orthonormal 9asis.
Example 3: Let w = Span {xi, X2} where X1 = (2, -5, 1) and X2 = (4, -1, 2). Cons~ct an orthonormal
basis forW.
Solution:
Let W = Span {xt, x 2} where X1 = (2, -5, 1) and x2 = (4, -1, 2).
Let Vt= Xt = (2, -5, 1)
-
Let,
v2 = x2 -(x2 .v!J v, I;
I (

Vt •VJ

X2 • X~ [' ,' Vt = X~]

= X2 - (
X1 •
X1 • X '. . ,

, , (14, -1, 2) • (2, - 5, ll\ (2, -5, 1)

= (4, -1, 2) - (2, -5, 1). (2, -5, 1))
8 + 5 + 2J{2 -5, 1)
(
=(4,-1,2)- 4~:1;1 (, 11\ I J
r t
3
= (4, -1, 2)-(1, 2, 2) = ' 7: 2)

=i(2, 1, 1) ·

Take v2' = (2, 1, 1).

 IIATBEIIATICS-D

Also, Ut = ,:.1,, (2, -5, 1) - ( :Ji - ~ ...1-.l

IYl ✓4 + 25 + 1 \-{is' --\J6' ...po) ·
and u2 = v~• - (2, 1, 1) = (ii .l... .1...)
llv ~I ✓4 + 1 + 1 \'VJ' -{i,' -{i, ·
Here {u1, u2} be an orthonormal basis for W.
The QR-Factorization Algorithm
mxn matrix with linearly independent columns ·e n A can be fa.CtOlceqi
mxn matrix whose columns form an orthonormal basis for Col
· invertible matrix with positive entries on its diagonal.

Example 4: Find QR-factorization of a mabix A where A = [ -i1 : ].

Solution:
Here,

a1 = [ - / ] and a2 = [ ; ]

Then,

Vt =a1 =[ -i1]
Vt 1 [ -1 ]
llt•,rvi11 =V2 1
Similarly,
;., .

-· ·rr ,. ' ,. ,
=[ ; ]-[- -i1 ] =[ : ] ~ :_'. ' I .•. - ,. I I J S

Vt 1 [ 1 ]
u2=llv2II ='V2- 1
0

Q = [Ut U2] ="2-

1 [ -1 1
1 1 _
1 .J

. 1 [ -1
R = QT A = 'V2- 1
1 ][ -1 3 ]
1 1 5 = 2
'V2 [ 10 1 ]
4
0
1
Example 5: Find QR-factorization of a matrix A where A-[~ 1
1
Solution:
Let the columns of A are x1, x2, X3. So,
Xt =(1, 1, 1, 1), X2 =(0, 1, 1, 1), %3 =(0, 0, 1, 1).
Let Vt =Xt = (1, 1, 1, 1)

Take, v2 = X2 -(x v!J

•
2
Vt
Vt•VJ
 Orthogonau....
- 0 .., lllld Least &quana .u;a \§w:#81 .: .
- co, 1, 1, 1i-(m. 1, 1. 1i- cu, 1, 1!l.I . --
(1, 1, 1, 1) • (1, 1, 1, 1)) (1, 1, 1, 1)

== (0, 1, 1, 1) -¾ (1, 1, 1, 1)
1
== 4 (-3, 1, 1, 1).
Set v2' = (-3, 1, 1, 1). Also,

== (0, 0, 1, 1) -(¾) (1, 1, 1, 1) -(122) (-3, 1, 1, 1)..

1
= 3 (0, -2, ·1, 1)

Set V31 = (Q, -2, 1, 1).

Thus, {vi, V21, V3
1
} be an orthogonal basis. Let {u1, u2, 1.13} be normalize of the orthogonal basis. So,
- ~ - (1, 1, 1, 1).
. u1 - 1lv1II - 2
V21 (-3, 1, 1, 1) .
u = llv2'II
2
~
V3
1
(0, -2, 1, 1)
u 3= llv3'11 = V6
Let Q .be a matrix whose columns are u1, u2, 1.13. Then
. i/2 '-3/{ii O_ ·
12
- j •

1;2 11-fu -2/-v 6

Q = 1/Z 1/{ii 11-{6
i/2 ·1;{12 11-Ji,
. ha A= QR by QR-factorization theorem~·Here, ·
. Smee we ve ' • . • , I

QTA= QT(QR) =QTQR =IR 9 R. I

Now,
R=QTA
1/2 1/2 1/2 ][1. 0
i l ~iii
⇒ R=[-3~~ 11..Jii 11""12 11-Ju
-2/'1°6 11-{6 111i, 1 1

3/2
21-Jv-
-1 ] .·
3/{12
0 21'16
 BXERCISE 8.4
The given set is a basis for a subspace W. Use the Gram-Schmidt process to produce an orthogonal
basis for W.

l. LU[!J 2. [~l[t] 3. [~],[-}]

4. [:J[~~] 5. [fl[~] 6. [1l[1]

7. F~d an orthonormal basis of the subspace spanned by the vectors (2, -5, 1) and (4, -1, 2).
8. Find an orthonormal basis of the subspace spanned by the vectors (3, -4, 5) and (-3, 14, -7).
9. Find the QR factorization
t

5 79] 2 3]
.
(1) [
A= ;
1
: (ii) A= [ !f •I

..
'1

{.

1.
LUGJ .
2. [OJ [5r : ·,
4 ,
2
4
-8
,
l ·
,.
L

·V
'

,_ ;_ THl!m -
-
• , I ,., . '· . '

4.
GJm 5. [!J [}l. . •
6.
[11 [}]
7.
[21ffe
1
r3/~r
-5J-{3D , 1.5/~
1;-{3o . 1.5/~
8.
[3/~] [1/V6]
-4/5'p.
11"'2
I 21{6
11{6
' I •
-
(

5/6 -1/6J ,
. 1/6 5/6 _ 6 12
9. (i)Q= -3/6
[
1/6 ,R-[o 6] .
1/6 3/6
-2/7 5/7]
.. 5/7
(u) [ 2/7
· 4/7
2/7
-4/7 'R
2/7
=[o7 J
 ; .. · ····-~
·•.:
,-,,-,,,

*_·-If:J;J .,

,;')-

Solution of Least Square Problem Ax = b

Suppose x satisfies Ax = 6. By the Orthogonal Decomposition_Theorem, the projection 6 has the
property that b - 6 is ort,,.og~nal to S:ol(A_},,- sol?:- ~ is orth~gomµ to ea~ J~?,lumn of A. !f a;~ any
column of A, then ai. (b-Ax) = 0, and a? (b-Ax) = 0. Since each a? is a row of Ar.
AT(b-Ax) = .o, I ••• (i)

Thus
ATb-AT.Ax=O
AT.Ax=ATb 'i

culations show that each least-squares solution of.Ax = 'b satisfies the equation
ATAx = ATb .. . (ii)
. tnX' equation represents a system of equations called the normal equations for Ax = b. A
Therna A .
tion of (ii) is often denoted by x . · . . .
so1u . Find a least-squares solution of the inconsistent system Ax = b for
1
Example • [4 . 0] . [ 2] ., ,
A= ~ i ,b= :1 .
Solution:
Here,
0
2
 IIATBBIIATIC8-II

and,

0
2

Since we have (by theorem 13) the set of least squares solutions of Ax = b coincides with the
non-empty set of solutions of ATAx = ATb.
Therefore,

... (i)
Here,

IATAI= 1117 11 = 84 .t: 0.

5
Th~

(ATAYl=~[~ -1J ·
17
Therefore (i) becomes,

t=~[~
The following theorem gives the useful criteria for determining when there is only use least squares
solution of Ax = b.
'f A' are linearly
es solution ~,
'
it is
'
and

Example 2: Determine the least-square error in the least-squares solution of Ax = b where

· [4~ iOJ ,b= [2]:1 .

A=

Solution:
Here,

ATA=[t 0
2
lJ[~ ~] =[17
1 1 ' 1 _1 . ~]
and,
2
1][ ] [19] .
A~b=[~ 0
2 .~ 1~ = 11 ·
Since we have (by theorem 13) the set of least squares solutions of Ax = b coincides with the
non-empty set of solutions of ATAx = Arb. .· .
If
I
Therefore,
... (i)
Here,
 Then, Ortbogona.uty and Least
~ Ill §i:i# '1

Therefore (i) becomes, ·

171] ·

~ ~ .!..[ s' . -1J[l9J - 1

Then,
. ~ 84 -1 17 11 =si[1:] = [~].
l • -

So,

b-A~=[~]-[!]=[~].
11 -3 >r 8
and,
,. . ' "
' • "j I

llb-A~H= ✓(-2)2+(-4)2+(8)2=VS4. Ir r
• I,
Thus, the least-squares error is ..../s4 . ., L):.
Example 3: Let A =
1 7 2
[! -: 7]
and; ~-=
, .. , _· ~5 -
[1].
'Find ~-least squares solution of Ax = b and
<_ J - •.
1

compute the associated least-squares_eri'or. · ~

Solution: u · s::.. ' , rrr, "'
Let

•' .
Then

Here

and ,

Then
=
3
83 28,
9 0
-65 -3] j •••

[~ 28 14

, r
1 3 0
83 28
J ,

-::] [Perfo~u
-65~_, ' I t I R1 ➔ 1/3 R1 .
[~ 2 1 -2 & ➔ 1/14
 <
· JIATJIEMATICS-D .
3
56 28
0
-1] -56 [ Performing J
rn 2 1 -2 R2 ➔ R2-9R
3 0
2 1 -1]
-2 J
[ Performing
rn 2 1 -:-2
R2 ➔ l/28R

~
3 0
2 1
-1]
-2 J
[ Performing
rn 0 0 0
RJ ➔ RJ-R

From the last matrix,

X3
2x2 + X3 = - 2 ⇒ X2 = - 1 - 2
3X3
X1 + 3x2 = - 1 ⇒ Xt =2 + 2
X3 is free · (".' being RJ: 0 = 0 which.is true)
For specific solution, choose x3 = 0. Then, the least squares solution of Ax =b is

f=[j =[-}]
and

h=Af=n t .IJ[-~J=[;J
Therefore the least squares error is
I\
11 b- b 11 = 11 (5, -3, -5) - (5, - 3, - 5) 11 = 11 (0, o, 0) 11 = o.

Example4: la A= [t 1l m. b = u = [~1J. and V = [~2J. Compute Au and Av, and

compare them with b. Is it possible that at least one of u or v could be a least-
squares solution of Ax = b? !\ r
Solution:
Here,
1 ] [ 5 ], [r
10 - ] I [ ,- ~ ~ Jr
Au=[~ 2 -4 -1 ·= -15 + 4 = ,-11
15-2 13
I

~, [8]
and Av=[¾ -41][5]
-2 [:10-2]
-15
2
-7- =
J5-4
:+- 18
-11
Here Au~ Av. So
..
IIAull = ✓s1 + 121 + 169-= V37i.
IIAvll = ✓64 +49 + 121 = ~ .··
This shows IIAvll < IIAull. Therefore, vis the-least squares solution of Ax= b.
 I SJ [4]
Ortboconauty lllld 1-at &qu.._
gsampleS: Let A== 3 l
'
[
-2 4 , B == -; , find (a) the Orth
~~

solution: (b) a Ieast-squate

'
I
• . s so Ution of Ax = b
~
ogonaJ projection of b onto Col(A) and
Let, I I . •
'
I

A= [u, u,) =[l '!] ~[}J an~

The~
Ut · Ut =: [ !] [-2
-2 ·
1
3 ]- 1
- + 9+4 =14
· 1

s.7 ps].lJ ·

U2 • U2 = [ !J l! = 25 ~ 1 + 16 = 42

~·- " ~
~:.i:.,i.J' • •
4 .... ~ -
> ·•J,I.? -i\11
,-.
~[-~J.[,~-]:=4•_;+,6a;,4~
-3 . -2
J ; I
~

, [4~:[s]
b.lJ.2:= f-21,
~ ,
.~:
1 "i'20,- ;2·- ) 2 ::i 6
-31., 4 ·
J • I •

Now, the orthogonal projection of b onto Col A is, -·

S = (! ·. :8 u, + (!,·.':.;) u, where A= [u, u~

1d
ffi ·.· . ~ 6~)UJ J~t[!t[,~~;1lf1~1 [il ,
Since, the least square solution
=
of Ax.~ b is the solution of A~ = ~- And we know,
C -

t-
b
/\ • Ut ·• Ut
Uu ·' - •'' ' · ··~-,
•· ;;'T"" ..........
,~ •• . f : ~' 1 J L

x = b . u2 · where A = [u1 u2]

[
U2. U i ·.

i.e.
/\ =[2!1/771j .
X .-7
:. I
I

I I '
•

ll /

llXllR(;ISI~ 8.5 ,..

I
'/I

. Find a least-squares solution ~f Ax =' b by i I,, ,. ,.

• . ~for~ and (b) solving fQr ~ I
1

•• 1-

(a), Cor~~j noff equatio ilie n A=P -~} b=[-:1 , r,,• ,,

I

-2 ..
I •,'

(iii) A = 01 i' 1
b = -4 . ._(ivt rA = 1 1· ' .. t O . .. . . ' ~ I

2 5 2
 r'
MATHEMATICS-II

2. Describe all least-squares solutions of the equation Ax = b.. ~

1 1 0
,,
1 1 0 2 ~'
1 l
(i) A-[~ nb=m
- 1
1
1
0
0
1 1 0

1 0 1
3
(ii) A= 1 0 1 'b= 6
5
~(
,1
1 0 1
3. Co~pute the least-squares error associated with the least-squares solution

A=U Ilb=[~l
4. Compute the least-squares error associated with the least-squares solution

5.
A=rn Hb=m-
Find (a) the orthogonal projection ~f b onto Col(A) and (b) a least-squares solution of Ax= b.

(i) A=[+ ~lb=[H . l1 i-1 -sJitJil.

.(ii)A=J! -
l~J
' I! . . '

I-
(iii) A =[i -1

6.

1. (i) (a) [-116 -llJ[x~

22
=[-4]
X11
(b) ~=[~]
(ii) (a) [1; :oJ G~ [-;4J =
(b) ~=[;]
"-[+4/3]
(iii) (a) [: ~]G~s[!] ' (b) X - -1/3 .

(iv) (a) :i][j=[t4]

[i (b) Q =[i]
., -:.

~ =[~]+x.[1] ~~(;}xtn
(ii) 3. ~ 4. l ~
2. (i)
I 1

5.
"
(i) _ b =3a1 + 21 a2 = (3,_-1, 4),,x" = (3,. ½) •
(ii)
"b = 32a1+ 3l ill= (3, 1, 4, -1), ~ = (23, 0, 3lu ,
(iii) " 13 314 5
b = a1 + 8) (13,3,3
a2-'.t"3= ~5,2,3,•3 , ~ •= 14 -~
1

6. Vector u can not be the least square ~olution.

 8°6 APPU«1ATI INEABOrthogonal!
MODE a..,.... tll IC,u.malll
~:::::::::=::~~~O~N~S~TO~ I.. ty and 1 -

A common task m . sc· IS -

among several quantiti
ience and engmee
. .
problem. a d of Ax es that v ary such isb\em y-re and und
vector andInste
y as observ .- b we write X~nng
- pro to anal
will amount toers™'.d the relationships
ation vector. - y and refer to X as d . solvmg a least squares
the data points w ~
esign matrix, " as parameter

If . ere on the line, the parameters ~• and ~• would satisfy the

Predicted .
y-value Observed equations
llo y-value
+ ~1(x1) =
\3o + ~1X2 !' = ,
I -

We can wnte
. this system
. lloas------:--~Y~n~_
+ !hxn

~=y,whereX{j]"·~=t~] ; Y=\;:\.
• ' , \ C • ,. LYJ
I,east-Squaies Lines ' ..
Suppcse ~ and ~ are fixed and consider a line
. Y = llo + !hX .- . \_ , ........ (i)-
,. ~ notation is commonly used ;nsteid of y ~ \l.x + c.J
as lll figure• Let (x;, Y1) be a paint- ' ·
Then (x;, ~• + ~ XJ) be a p0int on the Y o- Pol'"
line (i). In this1 case YI be obserVed - l"r Y1l
value of y and Po+ ~ x; be predicted l~<f+ ~,"11
1
·
y-value. Then residual is the' · . · '
Point on lln• / · · Reslduol

difference betweell 1 ob$1!r\11!<i •y- · --

value and predicted y-value.J
The Y =II.+~,•

~
least-squares line is the line (1) that -t--tx---x+-,;;.-----~----:.:,._~
1 x
the sunt of the squar!!S of ' •. •,
the residuals- lt is aJso called a\lin• of regression of y on x. 'Here ~., \Ii are called regteMion

-The
coefficients. . '' "·
)eaSt-squares JineS assunted that the errors in the data, to be acly in y-coordinateS·t
t

Ewople t: find the equation of 1east-squares line that best fits the data points (2. 1), (~. 1),

(7, 3) and (8, 3), .

Solution:
Let the equation be
 MATBEMATIC&-D

y = Jlo + Jl1x ... (i)

Then at the points (2, 1), (5, 2), (7, 3) and (8, 3), the line (i) gives,
1 = Y1 = Po + 2J31
2 = y2 = Po + SP2 ' ... (ii)
3 = y3 = Po + 7J32
3 =y4 = l3o + 8J32
In the matrix form of (ii),
y=XP ... (iii)
where,

The normal equation to (iii) is,

XT XJ3 = XTy ... (iv)
Here,

T
XX= [12 1
5
1
7

and

. T
X y= 2
[1 1
5 ( ,,. .' :. '

Therefore (iv) becomes,

,..
,(

⇒
, L I

Thus, (i) becomes,

2 Sx
,, y =7+ 14.
This is the required least squares line.
• 1 j . ..
The General Linear Model
Let J. '

.(f
... (i)
be the least-squares line. ff£ be a residue error defined by E = y - xp. Then (i) becomes as
. ' t •f
y = X~ + E ... (ii)
 Orthogonality and Leut Squana W fc¾#81 2
for p == (J}o, ~t) and X = (1, x) . ·- _
. .AnY equation of the form (~) is called a linear model.
'}1te least squares solution 'P cap' is a solution of th .
e norma1equations • •
XTXP=XTy. .
1,east-Squ~s Fitting to Other Curves
Let the points (xi, Yt), · · ., (Xn, Yn) lie on a scatter plot that do not lie close to any line. Let,
n
Y= L J3dk (x) ... (i)
k=O
wh~e, fo~ ··., fn are functions and Po, ..., J3n ar~ parameters. The line (i) is the general form of least-
squares line.
For a particular value of x (i) gives the predictor or fitted value of y.
The following examples show how to fit data by curve.
Example 2: Suppose the points (x1, y1), ••• , <xn, Yn) appear to lie along some sort of parabola instead of a
straight line. Suppose we wish to approximate th~ data by an equation of the form
y = J3o + li,.x + ~2X2
Describe the linear model that produces a least squares fit of ~e data by the
equation.
Solution:
Given equation is (')
y = Po + Ptx + J32X2 ~ I ••• 1
Let (xi, Yt) satisfies (i) with introducing Et that is r~s~due error ~tween the observed value Y1
and the predicted y-vahre. Then, 1 • .J • 1

Yt =Po+ PtXt + P2Xt2 + Et - ' 1 •.

.. . (ii)
S ~ l y at data (x2, y 2), ••• , (xn, y n) gives the value for (1) 1s, · .J

:=-:-:-:-::-:-::::: } : '
The equation (ii) and (iii) can be written as, 1 , :
... (m)

· y = J3X ~ E 1

:hae,y=[fJx{i" I :n~=[rJE{:1-
y
- ! '
I. •
' . i'
,, I

• I

_J:-----:~ 1
SurfQCe area
-----=-~~- .
X

.. of foliage
 MATHEMATICS-II

Example 3: Suppose the data (1, 1.8~, (2, 2.7), (3, 3.4), (4, 3.8), (5, 3.9) appear to a parabola instead of
a straight line. Describe the model that gives a least squares fit of this point by a
function of the form
y == Pix + PiX2
Solution:
Given function is,
y = P1x + P2X2 · ••• (i)
Let the points (1, 1.8), (2, 2.7), (3, 3.4), (4, 3.8) and (5, 3.9) appear. Then from (i),
1.8 = P1 + P2
2.7 = 2P1 + 4f32
3.4 = 3131 + 9f32 ... (ii)
3.8 == 4P1 + 16132
3.9 = 5131 + 25132
Since (ii) can be written as,
Y = px
where,
I• I I •
1.8
:.,,- } • !• I 2.7 I .

y= 3.4
3 ..8
,P=t~
3.9 ,., ' I J .

1 1
2 4
and X= 3 9 1',
4 16
5 25 -,

EXERCISI~ 8.6
Find the equation y = ~o + ~1x of the least squares line that best fit the given data points.
1. (0, 1), (1, 1), (2, 2), (3, 2).
.. -
2. (-1, 0), (0, 1), (1, 2), (2, 4).
/,

3. (1, 0), (2, 1), (4, 2), (5, 3).

4. (2, 3), (3, 2), (5, 1), (6, 0).

1. y=0.9+0.4x 2. y = 1.1 + 1.3x 3. y = --0.6 + 0.7x 4. y =-4.3-0.7.I

 Orthogonality and Least Squarea ID ICeij#BI

~~)1'NE~~B~P::R::O~D=IJ~CT==S=P~A~C~E~==~===~~========
Notations of length, distance and Orthogonality are useful m
, applications involving vector space.

v innltlon: An inner
V -associates product
a real . beon a vector space V is a function that to each pair of vectors u and
.- nµm r <u, v> and satisfies, for all u, v, w in V and all scalars c
' ,'. ,
i/·") <u, v> = <v,,, ·u'>· .. -, · ~ ' ' - .*·i~'
,1
• ~ , >,(.

(ii) ·<u + v, w> =:-<tt>w> + <~ w>i

' '

l •m• •")
, :<cu, _v ~ = c<u, v>. ,
_
, .\. , '

(iv) <ii, u> ~ Oai\d '<u, __ .

' ' •.- ' ,.~ ''. • .,.,-. : ; . : '.:'_'.@:f )''_, . ;'

'Av~r a~~· with an inn l;(c.tiis £all

Example_l: Let u =· (ui, u2) and v = (vi,. v 2) are in 91,2 and defined as <u, v> = 4u1v1 + Su2v2. Then
show that <u, v> defines an i,:mer product.

Solution:
Here,
(i) <u, v> = 4u1v1 + 5u2v2 = 4v1u1 + 5v2u2,= <v, u>.
(ii) For w = (wi, w2), · ·, ·_1 •; •• ,1 • •

,<u -'I-- v, w> =4(U1 + vf)wi•+ 5(u2 + v2}w2 • ·

-.. - - ) =- • ') (' '
t '· • ~ · · = iiu1w1 +' 5u2w2 + 4v1w1 + Su2w2

= <u, w> + <v, w>

(iii) f~~ any ~~ ft. , ,, · I ,t

<cu, v> = 4(cu1)v1 + p(cu2)_Y2,= c(4u1~1 + ?U2~2) = c <u, v>

(iv) <u, u> = 4u12 + Su_2 ~ 0

2

and <u, u> =O⇒ 4u,• + 5ui =Oif and only if u, =0, 112 =0 ⇒ u = 0
Thus, <u, v> defines an uu,er product on !IV. .' . '' .
Eumple 2: Let V be P, (polynomial of order 2) with the inner product
<p, q> ,s p(ta) q (ta) + p(t1) q(t1) + p(t,.)\, q(t,.) . . 1
"'-ct real numbers and for p, q e P,. Let ta = 0, t, = 'i, t, = 1, p(t) =
where to, ti, ti are di8 '"" ·•
12f and q(t) = 2t -1. Then compute <p; q> and <q, q>.

Solution:
Here,
<p, q> = p(ta) q(ta) + p(t1) q(ti),+ p(tz) q(t2)
 MATHBMATICS-U ,

= p(0) q(0) + p(½) q(½) + p(l) q(l)

= (0) (-1) + (3) (0) + (12) (1)
=12
and <q; q> ·= (q(to))2 + (q(ti))2 + (q(ti))2

= (q(0))2 + ( q(½) )2 + (q(1))2

= (-1) 2 + (0)2 + (1)2
=l+O+l
=2.
•~Difinition
•~,: --=--~---
(length or norm) ,
Let V be an irmer product spac ' . e ve'dor' V is a
, is d~fined as .}

Example 3: Let V be P2 (polynomial of order-2) with' the inner product

<p, q> = p(to) q (to) + p(tt) q(t1) + p(b) q(b)
where to, ti, t2 are distinct rea!, nuµibers _and ,for p, _q e P2. Compute the length of 1

p(t) = 4 + t and q(t) = 15 - 4t2 where.to= 1 and t1 = 2. Also, compute the distance
~ , ' ' ,} I 1. . , J

between P and q.
Solution:
For distinct real numbers to, t1, ..., tn and for p, q E P2, the inner product ls ·defined as
< p, q> = p(to)·q(to) + p(ti) q(t1)' + :..'."+ p(tn) q(t~) ; ' 1r ',
•r
Let p(t) = 4 + t, q(t) =15 - 4t2, to == 1, ti =2. \ I

Now - .' } tJ l

IIPII = <p, p> = (p(to))2 + (p(ti))2 I 'J ,l, ..... ·

= (p(1))2 + (p(2))2
=S2 + 62 =61. 1 ,... • I•

,
llqll =<q, q> =(q(to))2 + (q(ti))2
r ., ~r, ~

~' !: = (q(1))2 + (q(2))2 .. . I • • "

= (11)2 + (1)2 =Rl + t ~= 122

and the distance between p and q is,
IIP - qll = <p - q, p - q> = (p(to) - q(to)) 2+ (p(t1) - q(t1)) 2
= (5 ;- 121)2.+ (6 ~ 1)2
= 13456 + 49
= 13481.
 Definition (U
A
nit V9cto
r) Orthogonality and Least Sqwea W Jcmma 81
Vector With l
- - - -- f?ngthl,is'Un.it

==~=~;:::::::::::::::::::- -_J
Deff I · Vector. If u is an u
n tlon (Ortti . y vector then II u II is unit vector.
0
Let v be oSl nal)
< an.inner
u, v > :::: 0 ' Product spa
Th . ceanduvinVTh th
e Cauchy-Sch =;;~;--.:..._----'____
·_en
____ e_ vect-
or_s;...u_an_
d_v_are
___o~rth=ogo=nal
=-=i
if
Let V be an ~arz Inequality
Ulnerproduct
I space. Then for all u . V
Proof: If u == 0 th <u, v> I ~ llull llvll- ,vm ,
Then en the inequality holds, trivially. So, let u * o· Let y be a subspace spanned by u.
IIProju (v) II -- · ll<v, u>
<u,u>u II
- I <v, u> I
- lluU2 ·
V .
!lull ._ I<v, u> I
- llull
Since,
IIProju (v) II ~ llvll-
So . . _, " r j <;v, u> I . .
' llull s; llvll-
⇒ I <v, u> I ~ llull llvU.
This inequality helps to understand the following inequality.
,-
The Triangle Inequality
• f ,. ' ,, )

·' Let V be an inner prod~ct .sfa~:I · Then,for ~~.r·,u, ~ ~ V,

1
r

II u + vii ~ Hull + llvll,

Proof: Let V be an inner product space. Let_u, v•in V. : •
0

Here,
llu + vll2 = (u + v, u + v) = (u, u) + 2(u, v) + (v, v)
~ 11ull2 + 2l(u, v)I + llvl\2
~ UulF + 2liull llvll + livlP [By Cauchy-Schwarz lneuality]
(llull + llvlD2
= .. ,,

⇒ II u + vii ~ Hull + llvll• th Rl is an inner product space. Then show

For u, v in :J(Z, set <u, v> - .:nd the triangle inequality hold for x ~ (1, 2) and
- 4u Vt + S\12V2 en
Example 4:
that the Cauchy-Schwarz ,.
'J
y = (-1, O).
Solution:
Let x = (1, 2) and y = (-1, O). Th~= 4(1) (-1) + 5(2) (0) = -4 + 0 =-4
<x, y> = 4x1Y t + Sx2Y
 MATHEMATICS-D

Also,
llxll = <x, x> = 4xt2 + 5x22= 4(1)2 + 5(2) 2= 4 + 20 = 24
IIYII = <y, y> = 4y12+ 5y22= 4(-1)2+ 5(0) 2= 4 + 0 = 4
llx + YII = <x + y, x + y> = 4(x1 + y1) 2+ 5(x2 + y2)2
= 4(1 -1)2+ 5(2 - 0)2
= 4(0) 2+ 5(2)2
=0+20
=20
Now,
I <x, y> I ~ 4 ~ (~4) (4) = llxll IIYII
and llx + YII = 20 ~ 24 + 4 = llxll + IIYII•
Applications of Inner Product Spaces:
In this sub-sectio~, we arise some practical problems
Weighted Least Squares
Let Y = (y1, y2, ..., Yn) in ~n. Let
• A A A A
f 1, y2, ..., Yn are orthogonal projection of y1, y2,
. •
..., Yn
respectively. Let y = (y1, y2, ..., Yn). Then the sum of.the squares for error is
A A A A
= IIY - Yll2 = (y1 - y1) 2+ {y2 - y2)2+ ... + {yn - Yn)2. j·• '
SS{E) 1
•)

Let the errors in measuring yi are independent with means eq'1,al to~ ro and respective variance of 1

O'l. Then, the weights are

I ._:

i ..

Then the weighted sum of squares for.error is

A A /\ r .,, I
Ws.5 (E) = W12 {y1 - y1) 2+ wr (y2 - y2) + .'.. + wn2 (yn - Yn)
• Ii ~ t, t • .,~.,, • IJ . I 1 •
Sometimes, it is convenient to transform a weighted least squares ~roblem into an equivalent
ordinary least squares problem. I I , :•

LetWbeasquarematrixwith::gh[f on!dm~gJen-., ,._ :,.'

0 0 w f' /

• I

Then I

-~J[~:lyj- ~
[=w.yJ~~~]·..I .
W1 0
0
Wy= .
[
,; r p 0 .. . W .
., f i C
A.
w1y1
. A.
Wy = w2y2
A.
WnY
I , r
 Y
~ "1iii
I,et is to be constru Orthogon.u 8IUl Lout
And the measure of closeness
cted from the colwnns of ty
.IS thew .. . .
matrix A
...,_ Ill
etghled · Then Ax
i:-= 23
,._ II Wy - W ,._ error, = Yas close to
Let x is the 1
east-squ Y112 = IIWy-WAt.11' Y as po$ibl
e.
ares solution of the
W Ax_ equation

And the normal equationfor'-'-

- Wy I
me least•squares soluti •
f\A]AT OillS

Example S: Find the least-square . = IJ'{A T Wy.

, .. ) WAx .

~
3
(1, 4), and·
points (2' 3). Suppose
are greater
of the data
~ + 1n
s lmethy• =errors
fof the other 11,.x that best fil!I
. measuring the th • data (- 2/ h'
y-values ) (-1, 5), (0, 5),

Solution·. . · . •
poml!I. Weight thes e d ata halfo ast much thedata
e lastastwo rest

Let the equation be

.. . (i)
Y = ~o + ~~x
points
• (-2, 3):_(:.1~
- - 5), (0, S)r (1~4), ~ d .(2, 3): the line (i) .
Then at the
3 = y, = llo + (-l)II,.. , gives,

l 11
5 = y2 = ~o + (-1)~2 ... (ii)
5= Y3 = ~o + 0~2
i ql
4= Y4 = l3o + ~2 · \ ' .
;:r

3= ys = ~o + 2~2
1•
In the matrix £oi'ro of (ii), ... (m)

y=X~
-
where, 3
1 -2 5
1 -1 5
X == 1 0
1
y = 4 .11=Lt1· ,I

1 3 '
1 2
Given that the weight are u,ese data half as much as the rest of the data where y values of the
last two data paints are greater than.for the qthet Il°iftts- theref<l!'.", for th• weight function.

choose the diagonal entries 2, 2, 2, 1 and 1. Therefore,

2 0 0 0 0
O 2 0 0 0
W"' o o 2 O O
O O O 1 0
O O O 0
 IIATHEMATICS-D . .
Then
2 0 0 0 0 1 -2 2 -4
0 2 0 0 0 1 -1 2 -2
WX= 0 0 2 0 0 1 0 = 2 0
0 0 0 1 0 1 1 1 1
0 0 0 0 1 2 ·1 2
2 0 0 0 0 3 6
0 2 0 0 0 5 10
and Wy= 0 0 2 0 0 5 = 10
0 0 0 1 0 4 4
0 0 0 0 1 3 3
So ,,
I
-
1. , ., 2 -4
,... '1 ,rJ 2 -2
.rivx{(WX)=[! 2
2 1
-2 0 1 1] 2 0 -[14
- -9 ~]
1 1
1 2
and
6
1b
(WX)T (W)
y =[-42 , 2 2 1
-2 0 1 ~] 10 ·- -34-[5~]
4
3
Now, the normal equation for least square solution is 11

(WX)T (WX)P = (WX) (WY) I

i.e. [~! ;;] [~J =[ ~!]

This gives y

14f3o- 9P1 = 59
- 9Po + 25'31 = - 34
Solving we get
fio = 4.3 and P1 = 0.2
- .
• I
I
•
l y = 4-1.x
2
Therefore, the solution of normal equation is

, j
fio =4.3 and ~1 =0.2. Thus, the line is1
•J. .,, , ,[ • J( 1 ' .' I y = 4.3 + 0.2x,1 ° I -., i r .I ,'
- X

r' In cbnt:rast, the ordinary· least squares lines ' for ·these -2
data is • . r. -
y = 4.0 + 0.1 X
Trend Analysis of Data
0 l
ti
l
I

Let f represents a function whose values are known at to, ti, ...: tn. H there is linear trend in the
data f(to), f(ti), ... , f(tn) then the approximate values of f by ·the function of the form ~o + p,t .
 . . Ortbogonauty lllld
1,iJcewise, if there 1s quadratic trend in the da Leut ~ l a ~!
tJteform J3o + J}1t + Ji2t2. Let Pn be a pol . ta then ~e apProxunate val
. }'nOnuaI defined as i ues of f by the functio f
, < or p, q e p111 , no
p, q> == P(to) q(to) + P(t1) (t1) + , +
1,et po, Pu ....., Pn are orthogonal b · . q · ·· P(tn) q(tn).
as15 of p that .1
the polynol!Uals 1, t, t2, ... , tn. n are obtained by applying Gram-Schmidt
process to
And consider a polynomial g . p _.,,
m n at to, t1, ..., tn, is defined as ~-:
• . g(t) == Co+ Ctt + ... + Cntn .
where (Co, C1, ... , Cn) are m· Rn• com· :d .th· i j • rl
, c1ewi fle"be th , ..
· t,.g e orthogonal projection of g onto pn
A
g = <:opo + CtPJ + ... + CnPn•
Then gA.IS the trend function and . .
I I
' -
'
I -, ..,
Co, Ct, • • ., Cn are trend coefficients of the data.
Sthince po, Pt., .... , Pn are orthogonal which are obtained by the Gram-Schmidt process to 1 t tn
en, , , ... ,

<g, pi> = Ci <pi, pi> for i = 0, 1, 2, ... , n.

•• - I
_ <gi, Pi>
⇒
Ci - <Pi, Pi> r t • ..

1
The simpl~st( 'and lii{osf coknmon ..':..se+ of trend· Jmaiysis oc~s when the points
to, ... , tn can be adjusted so thatfthey ~e evenly spaced and sum to zero. Fit a
quadratic trend function to the data (-2, 3), (-1, 5), (0, 5), (1, 4) and (2, 3).
Solution: ' ·1 1 r) l ~ -=

Given data are (-2, 3), (-1, 5), (0, 5), ,(1, 4) and· (12, 3)~Let to, t1, ... , tn are distinct real numbers
then the polynomial Rn is defined as for p, q e Pn.
i~ \ <p,, q>: - ,p(~)1fl(f:p) ~ \ e• +, p(tn) q(t~). "-l , \ ' ' { •

Clearly <p, q> is inner product on Pn• ", , . .

1 5) (0 5) (1 4) (2 3). So, the polynomial 1s P2. Consider a
Given data are (- 2, 3), (- ' ' ' ' ' ' '
I mial 1 t t2. From these we observe ,
po yno ' ' t = (~Z, -l, o, 1, 2) (the'first value of the coordinate)
g = (3, 5, 5,.4,,3) ' (the second value of th~ coord,inaie)

Therefore,
Polyno~al: . 1 t t2 g
1 -2 4 3
1 -1 1 5
·5
1 0 0
vector value: 1 4
1, 1 3
1 2 4 5chmidt process to the
. that obtained by applying Gram-
Orth ogonal b as 15
Since po, pi, pi are · ,
polynomials 1, t, t2• Then '-
po= 1 .,. (i)
<t,~
pi = t - <po, po> po
 MATHEMATICS-II '
2
<t2, po> ) (<t , p1> :i ... (ii)
p2 = t 2 - ( <po, po> po - <Pt, p1>)Pt
... (m)
Then, p = Co po + CtPt + C2}'2

Where,
- <g, po>
Co - <po, po> .

Here, .
-=:t, po> = <t, 1> = (-2)(1) + (-1)(1) + (0)(1) + (1)(1) + (2)(1)
=-2-1+0+1+2
=O
<po, po> = <1, 1> = (1)(1) + (1)(1) + (1)(1) + (1)(1) + (1)(1) ·
=1 ~ 1+1+1+1
:,:: =5 I •

Then, (i) becomes,

Pt = t - (i)1 = t ~ 0 = t.
Again,
<t2, po> =;= .,< t2, l>L~ (4)(1} + (1)(1) + (0)(1), +. (1)(1) + (4)(})
0

( , ;:_ , ,I =4 +, 1.+ 0 + 1:+ 4 •>

I r I: n ,. . ,; fO } . ,-- . l

<t2, pi> = <t2, t> = (4)(-2) + (1)(-1) + (0)(0) + (1)(1) + (4)(2)

=0 t •. 1< .,: _ ,~.'Ji >?I ~ it • '-I . • I

<pi, pi> = <t, t>

' . +-(-1)(-1) + (0)(0) + (1)(1) + '(2)(2)
= (-2)(-2)

=4+1+0+1+4
' ;·,,:. ;)..., · · . =10 .· ,.!. .. • ' ,,

Now, from (ii} Pl =t 2 - (~}-.,Q= t

2
~- 2 . ·.
<g, po> = (3)(1} + (5)'(1) + (5)(1) +' (4)(1) + (3)(1)
=20
<g, pi> = (3)(- 2) + (5)(-1) + (5)(0) ·+ (4)(1) + (3)(2)
=-1
<g, 1'2> = <g, t2 ;.. 2> = (3)(2) + (5)(-1) + (5)(-~) + (4)(-1) + (3)(2)
= 6 - 5 - 10 - 4 +6
=-7
<p2, 1'2> =<t2 - 2, t2 - 2> =11t2 - 2112 ...
= (2)2 + (-1)2 + (-2)2 + (-1)2 + (2)2
=4+1+4+1+4
=14
 0rthogollallty aGd
Then (3) becomes ~
. .__~
~~
' I

' • • ' i., l§i.m,.~

\__s J Po+(.:!'\
P
+(~ •
lO)Pt
14) P2
• I

= (4)(1) _ (.!.'\ ( (1)

10) t)- - 2
Since the coefficient of . 2 (t - 2)
2
quadratic. P2 i.e. (t - 2) is not· extre:rnely _ '
small, so the trend fun . "
ction P is at least

X
-2 2

BXERCISI~ 8. 7 ··'
1. Let_ (a:z have the
. inner product defined as .<u' v> = 4u1v1 + 5u2v2 wh ere u = (u u 2) d
v- V1, v,) are m 912. and let x = (1, 1) and y = (5, -1). " an

(i) Find Hxli IIYII and I(x, y) I2-

(ii) Describe all vectors (z1, z2) that are orthogonal toy.
2. Let R have the inner product, define for two vectors x = (xi, x2), y = (y1, y2) in R2 as
2
<x, y> s 4x,y, + 5x y • Show that the Cauchy-Schwarz inequality holds for x = (3; -2) and
2 2

Y = (-2, 1).
Let VE P2 (poiynomial of order 2) with the jnnet product <p, q> = p(lo) q (lo) + p(t,) q(t,) + p(t,) q(t,)
Where lo, t1, t are distinct real numbers and for p, q e p2 and given by evaluation at -1, 0, and 1.
2
2
3· Compute (p, q), where p(t) =4 + t, q(t) =5 - 4t -
2
4. Compute (p, q), where p(t) =3t - t2, q(t) =3 + 2t ,
5 412
S. Compute IIPII and 1141~ where p(t) = 4 + t, q(t) = - •
6. Compute IIPII and 11411, p(t) = 3t - tl, q(t) = 3 + 21•.
 MATHEMATICS-II
2
Find the least squares line y = ~o + ~1x that best fits the d~ta (- , O)~ (-1, 0), (0, 2), (1, 4) and
7.
(2, 4), assuming that the first and last data points are less reliable. Weight them half as much as

. ----filWJL.2
the three interior points.
I
1. (i) llxll = 3, IIYII =fios, I(x, y) 12 = 225 (ii) multiple of (1, 4)

3. 28
4. -10
5. llt>II = 5'/2 , llqll .= 3-{3
6. IIPII = 2'/s, llqll = -{59
~ /
7. y = 2 + 1.Sx
· □□□

It •
'I •
1

..

')
ts

.; .

' ..
I.

' .
I i

.,GROUP "AND ,. I "0 ,

SUBGROUPS ,.

In this chapter we will introduce some of the basic ideas in abstract algebra. Our treatment is primarily
intended as a review for the reader's convenience. The readers
f \ t\I-
can omit
._
section• 9.1 and 9.2; and go to
'

next section if they are familiar with the contents of this section.

' ,, l

•·
,, I ' '
H

....

LEARNING ~UTCOME~_ ': I • J ' •

After studying tblS c11apter yoi should be able to: . '

f OperatiQns I
~ Sets and Se
• I

~ Mappings
~ Group
a Subgroups
 MATBEIIATICS-D

. .
The concept of set was introduced and the theory of sets was_founded by Russia born German
mathe~tician Georg Ferdinand Ludwing Philipp Cantor (1845 - 1918). The concept of set is a
fundam~tal concept for developing mathematical ideas. Here we provide a quick review of sets and
its operatlori&a- .
1n~m:elY, a ~is a well-defin~ collection or aggregate of certain objects. The objects that form a set
are called m~mbers or elements of the set. The sets are denoted by capital letters A, B, C, ..., X, Y, z
and its elements are ~enoted by small letters a, b, c, ...., x, y, z. A set with_finite elements is called a
finite set where as it:.is an infinite set when it .contains infinitely many elements. A set which
contains all the elements of the sets under consideration is called the universal set. A set with no
elements is called an empty set, null set or void set and is denoted by cp.
e
Example l: The set B =(real solution of + 1 =O} is null set since
x2 x2 +1 =0 has no real solution.
This equation has imaginary solution x = ± i, where i = -J'=i. is an imaginary unit.
Example 2: The set A =(1, Z 3}, B = (a, e, i, o, ut; C =(x : x > -2, where x is an integer} = (3, 4, 5, ...},
D = (x: x2 - 4 = OJ.are examples of the'Set. ' ., . .
Here,
A, B and D are futjte sets where as"· C. is an infinit~. ~t.' Th~ ~versal set in this case is
U = {a, e, i, o, u, -2, 1, 2, 3, 4, 5, ... ,} ~ I

Au"B={x:.re Aorxe B}.

Example 3: When A = {a, b, c, d} and B = {b, g, h} th_e n
A u B = {a, b, c, d, g, h}
Intersection of two sets A and B written as Ar.B, is defined as the set of all elements that belong to
bothAandB.
AnB = {x: x e A and x e B}.

Example 4: A= {1, 2, 3}, B = {2, 3, 4}, Ar. B = {2, 3}

Complement of a set A with respect to its universal set U is the set A' of all the elements of U that do
not belong to A.

A= {x,e_ U: x EA}.

In example 3, complement p( B is B' =~{R, .c, d} and.complement of A,is A'_-= {x ~-U :.x ·
Two sets are said to be equal if and only if they consists of exactly the same elements. The eq~al sets
must contain each other i.e. A = B ¢:> A ~ B and B ~ A where ~ indicates two way implication or
logical equivalence.
f ~ ) • •

The above mentioned set o·p eratio~ support f~r the formation of set algebra over subsets ·of a
universal set U which satisfies certain laws that can be verified · either by using the concept of
eq~ali~ of sets br _by using the Venn diagrams. Foll~wing laws are true for anr 'su~ts of U: ,
Idempotent laws: AuA = A, AnA.= A
Commutative laws: AuB = BuA, BnA = AnB
Associative laws: Au(BuC) =i(Au)uC, M(BnG) = (Ar\B)(>)G
Distributive laws:
Au(Br\q = (AuB) n (Auq 1

. )
An(BuC) = (AnB)u(AnC),
Identity laws:
· -A,A U=U,An<1>=<1>,Ml1=A · · ··· ·· ' . 'd · · . kno~
~ uci, ~ u . (1806-1871) first introduced following 1 entities
English mathematician Augustus de-Morgan . , , . .
as De Morgan's laws:

-(AuB) = An B. ''·

A-{BU9 = (A-J3)n(~-Q,
' ' · · ) ( d 'A cross B') is
A-~Q = (A - B) u (A-C . sets A and B is denoted by AxB rea
,.... . . product of two . ,
The eartesian
Cartesian product: A b e B}.
. d as .AxB = {(a, b): 'v a e ' 2) (b, 1), (b .
d efirie thenA = ' '
- b c}, B = {1, 2},

· to b. ~ is said to.be an eq
.· b t R of A xA
tion: A su se
Equivalence ela
properties hold: A (Reflexive)
R for all a e . try) -
a. (a, a) e R::::) (b, a) e R (Synune) e R (fransitive)
b. (a, b) e d (b c) e R ::::) (a, c
c. (a, b) e R an '
 MATHEMATICS-D , · .:.:> : - • • ••
1
A subset of AxB is also taken as the binary relation in A and write (a, b) e R to mean 'b is related to a.1
Congruence modulo n: Lef n > 0 be a fixed integer. The~ a is said to be congruent-to b modulo n and
~ ~;;t 5 b(mod n), if n divides (a - b). Here n is called the modulo of the relation. ,
E~a~~le 7: 83 = 2 (mod 9) since 9 divides 83 - 2 = 81. Similarly 83 = -2 (mod 5) since 5 divides
83 - (~2) = 85.

Mappin~ . , .
A mapping or a function/ from set X to set Y is a relation or rule that associates each element of X to
unique E;lement of Y. A mapping from a set X to another set Y is dentoed by
f : X ➔ Y. Here are the lew types of mappings, injective, surjective and bijective mappings.
'J . 1 )' ' .,
Injective mapping: A OI~e-to-one (injective) mapping/ from set X to set Y is a mapping such that
each x in X is related to a different y in Y i.e. distinct elemen~ in X has
distinct images ht Y. We can
restate this definition as either/: X ➔ Y is one-to-one provided ft..x1) =ft..x2) · implies x1 =' ~ ... • ; ·
or / : X ➔ Y is one-to-one provided Xt =#. X2 implies f (x1) '# f (x2): . , r:
Example 8: The mapping/ : I ➔ I is given by/ (x) = ·3x +· 7 is one-to-one, since :f(a)#=f(bt means
3a + 7 = 3b + 7 ⇒ a = b.
Example 9: A function/: I ➔ I given by /(x) = x 2 is not one-to-one since we h~ e f(3) = f(-3) = 9,
but 3 ~ -3. So the.definition of one~ne mapping is violated. ·
De.f!nition: A funai ~~
»;.-i,
X such that.I{~ - l ·
J l _.,.,,

I . - I
This can be restated as: A function f is onto when its co domain equals to its range, i.e. if f: X ➔ Y
then .f(X) = Y. . .. ' .. , . , n ,
: · ,, t ' -.'rt6 • ol:
Example 10: A function/: I ➔ I given by /(x) = 3x - 7 is onto. If for every y in I, then (y + 7) an
+7
T Y..:!:.Z
are also real numbers. So, x = y
( ~ - (YL!:\ - --
3 e I and f (x) = / \. 3 ) = 3 \: 3 - ) 7 - ,),
hence for any yin I, there exists an x in II such that/(x) = y. .·. -, "' ~ ( ~, ., , :

Example 11: A function/: I ➔ I given by /(x) =~ is not onto since for Oinlthere is-no x ml such
,-1 l .~ I

that /(x) = x 0.
1
= r. • l, I • _4" \ I (;_..)

One-to-one correspondences: A mapping is called a one-to-one correspondence, (bijection) if it .is

both one-to-one and onto. .
, ' I • ... • , .. flt

Composite mapping: Given two functions f: X ➔ Y and g :. Y ➔ Z, then _the composition of

· mappings f and g is a mapping gJ: X➔ Z defined as (gJ)(x) = g(f(x)) for x e X.

•
 Example 12: Find the composite ma • Group and Subgroups Ill E#ri-91
f(x) = x2 + 5 and - PPtng of two functions .
. g(x) - x + 1 respective! / . X ➔ and g ! Y, ➔ Z defined by
Solution: · .. y. Al~~ ~eek. whether ,\60&/
,~ "(x) = (fog)(x) or not?
The composition of mapp;... / ,, , ... , , , ,
&ugs and g given by I ,)• ")
(gc/)(x) = g(f(x)) = g(x2 + 5) = (x2 + 5) + 1 = x2 + 6 , ., : .•. A
. . . ..,'-' l. :t· ..:-··

(fog)(x) = /(g(x)) =f(x + 1) = (x + l) 2 • •• ) ,, , ,, .•

+ 5 = x2 + 2x + 6 'l"'

Thus, we have (g0j) (x) ~ (fog)(x). • __ _

Well-defined • r c ' ' · - 11 , '
function: A function/· X-+ y · •
Inverse image of an · ISsaidtoJ:,ewell~efinedifforalla,b,~ X,a=b ⇒ft.a)=f(b).
e1ement Suppose/. X y . .
ele~ent y e Y with respect t - . : -+ is a given mapping. Then the inverse image of an

,• . ·
I
image of y = {x: y = .ftx)} In mapping/ 15 _the set ~f elements in X whose ~age is y. That is inverse
xample-9, the inverse rmage of 9 is {3, -3} since 9 = ft.3) = ft.-3).
Inverse mapping: Hf: x-+ y is t . ~ •L•.. · ,. -.. ·,
mapp· gfrom y to X
J ., , ..

. .. . a one- o~:i;te and onfo (1.e. b1Jective) mapping, then we have a

called inv . · · . . .. that ~~~iates ~ach element pf Y to uru~ue' element of X. S~ch ~ mapping is
erse mapping of/ and denoted by r-1. 1 . · - , ·,· ,
cr • ~- • , - ,·
1
~xam~I~ l3: Find the inverse fun~on ~f the function/: I -+ I ~e given by /(x) = 4x - , 1 = y. ··
l • ~ : l ,:, •n11 •

x
h •',

Now, interchange and y _a nd solve 4y -1 = for y to,get•y..= x x: ~- Thus;j-t (x) = x: 1

.

<tomplex Numbers: K. -sef of real numbers· geomemaally represent a ~traight

line: called x-axis. A complex number can be regarded as ·a point in 'the
Euclidean plane, as shown in figure *hich is generated by the x-axis (real
axis) and- xy~axis called imaginary axis. The set of ·compJex number is
denoted by C and define by . . real axis
C = {x + iy I x, y e R}
every
..
real numbe~ x.caz:i
.
~ ~ttenas
•
x ·=. x·'+ i 9' he~ce is a complex nwnper i.e. •every
\ ' It ~
real
t
number
"' I
is J • ,I

a complex number. Complex numbers we~e deyelo?:d ,~E:f ;the ~evelop111~n~ ~£1real numbers. The
complex number i was invented to provide a solution to the quadratic equation
x = •- 1. So we .requrre that'·2
•-2 I •··• ri
1 = -:1. . • •
,. ·, '
· •
I •
.
,.•
.
•
.
. I I . . . 'I

J,. ; . ...• t
• r
Unfortunately, i has
•-
been > '
called an imaginary number, and this terminology has led generations of
,.Jj I '

students to view the complex numbers with more skepticism than the real numbers. Actually, all
numbers such as·l, 3, 1t, -V3 and i are inven1ioris of our nuhds. I
I . I ~ I . - ' ,

Multiplication of Complex Number

Let z =a + ib and w =c + id are two complex numbers then their multiplication is zw = (a+ ib) (c +.id)
= ac + a id + i be - bd
rJ, . I:' '
• I

..
,
= (ac - bd) + i (ad + be)
r
Exampl~ 14; Compute (3 + Si) (6 + 7i) ·-·
= 18 + 21i + 30i - 35
'' (

. - = - l 7 + 51 i · • . · ·tude or distance of complex number z =a+ 1·b fr om

The absolute value of complex number i.e. magru
origin is given by Iz I = {a2 + b2 ·
 MATBEMATICS-.JI -

The polar representation of comple)c·number R

z = r (cos8 + i sin8) =te1e where eie = t:os8 + i sin8
if z = r1 (cos81 + isin8i), W = r2 (cos 82 + i sin82)
Then, zW = r1 (cos81 + i sin81) r2 (cos82 + i sin8i)
= r1 r2 [cos (8i + 82) + i sin(82 + 8i)]
and

Similarly, zn = I"' (cos n8 + i sin n9) for any integer.

Binary Operations: A binary operation * on a set S is a function mapping S x S int<;_> S: Given, for
each (a, ~)e S x S, we will denote.the ,e lement* ((a, b)). o~ S by a * b. Intuitively, we may regard a
binary operation* on Sas assigning, to each ordered pair (a, b) of elements of S ~ S, an element a• b 1
ofS. - .
f t •: r I• ~ • l • I

Example 15: Usual addition'+' is a binary operation.on the set Ill. Similarly usual multiplication'.' is
also binary operation on I. Similarly Ii can be replaced by C;. z-,-1+, or z+.
Example 16: The ~t M(R) be-.~e set of ~ m_atrices with, real ~~tries. The usual matrix addition '+' is
not a binary operation of this ~et. S"'<:e A + _B is no~ defined for all ordered pair (A, B) of
matrices ha~g _d ifferent number of rows or. o~ ~olumns.
---~:""'l!-.~-
BefinIllon:

Remark: From the definition of a binary operation* on S, th~ set Sis closed under'*, but a subset may
not be, which can be seen in the following example. · - ' ,.-, ·
. '·
Example 17: Addition is a closed binary operation on set of real nu~ber R, ~ut for_R~~which is_set of
a non zero real numbers, addition is binary operation but not closed in R*. Since -2 er,
• ' I - .. .. '· •

· · 2 e i* but 2 + (-2) = 0 E Ii*. · ' ·

Example 18: Set'+' and'.' be the usual binary operati~ns of addition and multiplic~ti~~ 011; U!,e set
z•, and let H = {n2: n e z•J. Determine whether H is dosed under (a) addition and (b)
multiplication. ·
Solution.
Here, for addition choose a, b e H then a =m2, m e z• and b =r2, r e z• then a + b = m2 + r2 may
not be the perfect square.
Taking a = 32 and b = 22 then a + b = 13 which is not a perfect square. So, H is not closed under
addition.
· For the multiplication: Taking a= m2 and b = r2. Then a. b = m2. r2 = (mr)2 where m re z•.
Hence '.' is closed binary operation in H.
 Group and Bubgrou,- w ~
pie t9: Let F =-.ff : f is~ real valued function . - ~ , 28
~ •
'+'' ,_L
,
, •t and -'o' !to'- "'I d.
ou.,c, ose Iit F. ,
haVIng domain I} then all the b'
mary operation
soJution: . '01 ' '

Since ¥ f, g E!. and ¥ ,x e R we have

• I
(i) f + g by (f + g} (x) = f(x} + g(x} e F
(ii) f - g by (£ - g} (x} = f(x} - g(x} e F .,
(iii) f. g by (f. g) (x} = f(x} g(x) e F
(iv) fog by (fog} (x) = f(g(x)) e F

Example 20· On z+ wed

• ,
fin b' ·
e e a mary operation * by a • b equa to the smallest of a and b, or the
· 1 i ,,
common value of a =b. Thus, 2 • 11 =2, 15 * 10 = 10 and 3 * 3 = 3.

Example 21: On z+, we define a bma,y operation •' by a •' b = ~- Thus 2 •' 3 = 3, 25 •' 10 = 25 and
5,., 5 =5.

Example 22: On z+, we define a binary operation •" by a•" b = (a•b) + 2, where* is defined in
example 20. Thus 4 •" 7 = 6 =4 * 7 + 2 =4+2 =6, 25 •" 9 =11, and 6 •" 9 = 8.
, be S.
,ce S.
- -
Example 23: Determine the binary operation • defined on z+ by letting a * b = 2ab is commutative
and associative or ·not. ~ -.
Solution:
. '
Given,
+
a • b = 2ab for all a, b e Z
for commutative
for any a, b e z-t: we have
a * b =2•b =2ba =b * a ( ·: ab =ba)
\,. • I I ... . ·1

H ence, z+ satisfy the commufative undet "' .,

., . t ,. •• r·. , ,. P I L .•1 •
.. ,,.. ... ' !

For associative .. ,

1or
£ any, 3·, b, c, e z+ we have to show ) I

(a * b) * c = a * (b * ~) ' l ' I

j•r ,-,,r J'

. Since, (a* b) * C = 2•b_*, C = 2rb.c '1 •
f" r

' 1(
2~ . '
.: . ' a * (b * c) = a • 2~ = 2a.
• I I ·, •
Since, a * (b * c) ~ (a * b) * c f' ·
. .
So, it is not associative. d to be if and only if statements. Theorems are not always
Remarks: Definitions are always understoo ti is ever used for theorems.
if and only if statements, and no such conven on
 MATHEMATICS-D ,

E:xampl~ 24: Usual addition of real numbers or complex numbers is commutative but subtraction
and division are not, similarly, addition on the set of all 2x2 matrices is commutative
but multiplication is not commutative binary operation.

Example ~5: Determine the binary operation "' on Q defined by x * _Y = x; y is associative or

commutative. [2076]
Solution:
Given,

x *y = ~
3 for all x, y e Q
Associative: for any x,.Y, z e Q we need to show x * (y *z) = (x * y) .* z
Here, x * (y *z) = x * (Y; z)
. x+ ~
. l
3
= 3
,_ 3x+y +z
- - 9
x+v
',,/ (x * Y.) * z =3*z .
~
3 +z
=
f.
3
= X + V + 3z
9
Since x * (y * z) -:t: (x * y) * z so it is not associative.
Cumulative: For any x, y e Q we need to show x * y = y *x
Since, x * y =~
3
·
., J

~ ( ·: X + y- = y + x)
= 3
=y•x
:. · It is commutative.
Hence, the binary opera~on * on Q is commutative but not associative.
Definition: A non empty set with one or more binary operation defined over it and satisfying certain
law of binary operation is called Algebraic Structure or Algebraic System. For example,
}
(N, +), (Z, +), (Q, +), (R, +), (R, ·) etc are algebraic structure.
,,"" L
Isomorphic Binary Structure
Let us consid~r a binary algebraic structure (S, *) to be a set S together with a binary operation * on S.
In order for two such binary structure (S, *) and (S', *') to be structurally hlike in the sense we have
•
described, we would have to have a one-to-one correspondence between the elements x of S and
elements x' of S' such that
ifx Hx' andy+-+y' thenx*y+-+x'•y1 ,., ...., .... (i)
 . Group and Subp,upa a, §w.iia
J. one-to-one correspondence· @xists ,if t h ~
iJJlPortant to descri~ a on.~to-one,corresp:ns:::e,and S'! have the sante number bf eletttents; It i)
5,· for such.gha function, we regard th . by giVmgiacon~1o-one function"' .
d e equation «l>{x) = x' d" · ,,, mapping S into
the Jeft-to-n t or er. In terms of"' th final as rea mg the one-to-one pairin '.
. S' . th ,,, e f4 corresp d . . gxH x m
strllcture m is e same as in S, can be ·, ., on en: e m (i), which asserts the al b .
. . . expressed as «I> (x• ) _lci,(x) *r (y). . , _ge ra1c
s~ch a function showing that two al .
i50morphism. gebraio systems are structurally alike, is<1 known as an

ilffom LetjS, *) and ,( , •') be

ed:isotnoiphism. if
(i) .
«I> •18 a
Ol).e;-~o-one ~ction
(ii) ti) is onto .-: •,·,. 1• ··'::' ,.. • .J.

(iii)«!) _is ho~o~oiphism i.e. cp(x * y) = cj)(x) *' «i,(y) for all x, y e S . . -. - - -
In ~ ~se s and s•' are isomorphic binary ·striicttttes and denoted l>j s "" ~·- . T• ii.- ·,·. . ~ _r, I l

- In fact an . hi .. . .' . . . r • • • • '' • J ••., n r_ •

• , T ' . 1$pmorp sm 1s a ~rw-tq~o:r:ie and o.r:ito.ma}?ping,1which preserves ~tru.ctures.

1 11
In a
1
b ~ algeb~aic structure "the structure" is the binary oper~tion. · • '' •

Consider an ExamplE: '' 'j • j(J

• rl
'·+ 0 1 2 '* a x y z
0 0 1 ••r l ,,.J
2 a a b C X .. X ' z
, . '2 . .. . b ;, · t;,.rf y y
.. 1 '. ',.1 0 ~ .... 1" J • c· a .•'l! _l ~ X

2 _ 2_. 0 . J _. .:.. c_ . ,)f ~ ... -., . ~ ·, ~:~.!- b ··z y x z

Notice that the structure of operation + on {O, 1, 2} is the same as the structure of * on {a, b, c}. This
can .be seen ,repla~g 9{ 1;,~ .with .a; .b, <; (re~pectively) and--:4" ,._with*. _Then any equation involwg the
first table yields an equatioi;i _involving the ~ <>\\~ tap}~.. (an~ ~ c~a-v~rs~).._ $o the,only ~~ence is
binary operation ~ d b!Jta:iy o~ra~o~ ~ ~ ~ne ·of noEtfio~ ~ the fu:st ajid_second tables repr~sents
isomorphic binary operations (on the appropriate sets). However, th~ _tl\ird. table is~fup.damentally
different. Since the bil)ary operations when _a p~_lied !~~ :pa~ °if ,!hs.~ e Je~e~ents _~ eld that element
(it is an idempotent binary operation). This is ~ot the case in t1h'.e 6;t'st two tables and so •' is not
• " • · • . ' ~ ... ' ,,. ... _ j I 1 ' ' _..
isomorphic to + nor *. J
. '
Example 26: Let us consider a binary ~truc~e (I+)_w!th _opera~~n the us~al a~ditio~ an~. the
n structure (ii+,.) where'.' is the usual mulnphcation. Tlien a function cl>: I ➔ I+ d~~mea
by q,(x) = e" is~ isonio~~~ P)- , .,. , 1, ,, ;>! · ,. " 1 , ,._ "

Solution:
H cj>(x) = cj>(y), then e" = eY ⇒ x = Y· l • I ·l I • l • 'f.)

cl> is one-to-one _ . l .:.

- "E R+ R+• X = lny E R such that <I> (x) =~( lny) =~ny =y.
For every y y - e ' · ..
q, is onto. . ·
. have "'(x + y) = ex+y = ex eY = q,(x) . <l>(y).
Ill) L.., l • ,,
For x, y e 111,, we "' .
q,(x + y) = ¢>(x).cj>(y} which shows there cl> is homo~o~hism~
. d /1U) +) is isomorphic with (I ' .).
Thus, q, is an isomorphism an \IIW'
 ~THEMATICS-II , ~
-
Example 27: Let (Z, +) urd (2Z, +) are two binary structures, where Z is the set of all integers. Then
· the function ♦: Z ➔ 2Z defined by ♦(n) = 2n is an isomorphism. ,
Solution:
,t ••
li ct,(m) = ct,(n) then 2m = 2n. So m =.n. 1Jtus ♦ ~s one.to one. l.

li n e 2Z, then n is even. So n = 2m for m ~ i e z. Hence ♦(pi) =~i) =n. So ♦ is ·onto·2Z.

let m ,: n e ~- Th~ equati~n «!)(m+n) ~ 2 (m + n) = 2m + 2n = «!)(m) + «!)(n).
.t
ct, is homomorphism.
Then ci, is isomorphism.
•> tie 'a: •

In particular for usual addition binary operation defined in the set of real or complex numbers, 0 ~
identity element. Similarly for usual multiplication binary operation defined in the set of real or
complex numbers '1' is the identity element. For the matrix addition binary ,operation defined on
~~ ma~[~' ~] is the ~d~~~ ~d fo~ the1

set of all 2 x 2 ~trices with real .or complex entries,

multiplication binary operation defined on the set of all 2 x 2 matrices of real or complex ~ntries the
matrix[~ ~]is.the identity element.

Th•llN•" (lfflj.q
eiement i.e. u th!

1. 1
Closure: ¥ a, b e G then a•b e G.
2. Associativity:¥. a,·b, c ·e G then (a* b) * c =a* (b ~ c). . ,,.
3. Existence of identity element e: ¥ a e G there exists an element e e G.. Such that;
a• e = e *a= a.
~r,
4. Existence of inver~ element: ¥ a e G there exists ai e G such that a * ai = e = ai ~ a.

Example 28: The set of real numbers R forms a group under addition dperation. ·
Solution:
i. Qpsure: for any a , b e R, a + be R
• I
. (

. 3 3 -17
for example: If a= 4, b = - 5 then a+ b = 4 -.s = 4 e R

ii. Associative: for any a, b, c e R we have,

a + (b + c) = (a + b) + c
for example: a = 2, b = - 3, c = 5 then
a + (b + c) = 2 + (- 3 + 5) = 2 + 2 = 4
 (a + b) + c =(2. - 3) + 5 =- 1 + .5 = 4 Group and Subpuupa Ill §R 9]
So a+ (b + c) =(a+ b) + ~ -, '- ·
I • _If'
iii. Existence of identity: there is
an e1ement Oe R h th
0+a =a +0= a sue at for any a e R.

for example if a= 5 then 5 +.o = 0 + 5 = 5

iv. Existence of inverse: for each · .
a e R there _exists an element - a e R s t
a+ (- a) = (- a) +a= O ·

for example if a= - 5eR then - a= - (-5) = 5 e R

and (- 5) + 5 = 5 + (- 5) = o
so every element of R has its inverse in R
so (R, +) is a group.
I I

s Example 29: The set of real number R is not a group under multiplication since if satisfy closure
· associative and existence of identify but it does not satisfy existence of inverse ·i.e., we
· · · · · of zero. ·
, · · · utativE:- i.e., A group
. a,
e
Example 30: The set of integers Z is a group under the binary operation addition. But the set Z+ under
addition is not a group; there is no identity.ele~ent for + in ,r.
Example 31: The set of all non negative integers including O (zero) under addition is still not a group.
ilii There is an identity element 0, hh1 no additive inve~ for every element. .
Enmple 32: The familiar additive properties of integers and of rational, real and complex number
show that Z, @, I and C under addition are abaian group.

Eumple 33: The set z· (set of all positive integers) under multiplication is not a group. There is an
ide~tity 1, but has no inverse elements of 2, 3, 4 etc. .
• The familiar multiplicative properties of rational, real and complex numbers, the set Q+
' Example 34. .. be and the sets Q* 1· and C* of non zero numbers under
and R+ of positive num rs '
multiplication are abeian groups. ·
d functions with domain I under ~ction additipn is a group.
Example 35: The set of all real-value
This group is abeian, . ·· .
. . d matrix addition is a group. The mxn matrtx
. . (Ii) of all m><n matrtces un er
Example 36: The set Mm"n . . trix This group is abelian.
. o is the identity ma ·
' with all entnes , . ultiplication is not a group. The nxn
><n matrices under matnx m .
bample 37: The set Mn(i) of all n . verse
· QhasOOUl · ·
matrix with all entnes nxn matrices under matnx
0
f all invertible
m) consisting
Example 38: The subset S of Mn \ ..
multiplication is a group.
 Example 39: Let• be defined on Gt by a• b =a:, then show that Gt form a group under the b.
mary
operation•· [2078]
Solution:

Oosure: For all a, be @+, a*b = a; e @+. Hence elements of ar are closed under*
Associativity: For all a, b, c e @+,
ab abc
(a• b) •c =2•c =7. }

. 5
A gain, a * (b * c) = a * 2be =4abc
e
Thus, * is associative
Existence of Identity:
=
Now,
1
2a
and 2•a= 2 =a

Hence 2 is the identity elem~nt for *.

•I
Existente of Inverse:
., • • j•
Finally a* b = 2 = b * a, where bi~ inverse of a under·•.

ab ba
2= 2 =2
4
b =-e Q+
.a
. , ·n.

l!

4 . .
ai = - is an inverse for a. Hence Q+ is a group under the binary operation *·
a
r'
Elementary Properties of Groups
Theorem: If G is a group with binary operation *, then the left and right cancellation laws hold in G,
that is, a* b =a* c implies b = c, and b * a_= c * a imp~es b = c for all a, b, c e G.

Proof: Suppose a* b = a• c. Since G,is a group the~ for each element a e G there exists a~ _e G such
'I
that:

ai * a = a * ah= e

Now: a•b=a•c
 ,ring ai both sides from the left
,plr . ,
1
81*(a* b) = a * (a* c)

~ the elements of G are associative with the ~.

. · mary operation *. Hence,
(a1_* a)* b = (a1 * a) * c

plies: e * b = e * c

b=c [B
y using the definition of identity element]
~gain, for the equation: b * a = c * a.
5ince G is group with resi,ect to a bin . · ·
. . . ary operation •: Hence for each element a e G there exists an
element a"EG: a* a 1 = a 1 *a= e.

Now applying a' from the right of the equati~n, b * a = c * a

'
⇒ (b * a) * ai = (c * a) * ai

By using associative property:

, · ft9t

b * (a * ai) = c * (a* a 1)

By using the definition of identity: b = c.

.. £G, then the

Proof: Given that G is .a group with binary operation•; for the equations a* x = b arid_y * a= b where
a, b e G, we need to show that the solutions x and y can be obtained in G with their unique values.
Now
a*x =b · '.... (ii)
Since G is a group with a binary operation * and a e G then there exists ai ~ G. Such that
ai * a = a * ai = e . (' ,. f I

Now fr9m (ii): a,i * (a,* x) = ~i b *

· ,By-associative law: (ai *a)* x = ai *·b
⇒e * x = ai * b
By using definition of Identity . · ,·_ · '· ·
. x 7 ai * b is solution of equation a * x - ,b. . .
· · if ssible there is an element x1 e G is another solution of
Now for uniqueness of x; suppose po . ··
. ' '
equation a * x = b .... (m)
then, a *x1=b
Now, from (ii) and (iii), we have
a• x =a* x1
by using left cancellation law:
x == xi; thus solution is unique.
 'JIA'THEMATl~D

In the similar fashion, y •a= b has unique solution b * ai i.e. y - b,. ai. .
TMorem: lri a group G with binarv ti
- - J opera on*,
th ·
ere IS only one element e in G
e * x • x • e = x i.e. identity element in a group is unique. sucli ttii
Likewise
• foreach a e G, there IS• onIy one element a1 m
· G such that a• * a = a * a1 =
mverse element of a group is unique. e i.e.
Proof: Suppose if possible there is an another identity element ei e G: e1 * x = x * ei = x.
Then, e * ei = ei supposing e as identity element.
Again, e * e1 = e supposing e as identity element
Hence, ei = e
Turning
using :t hetod the uniqueness
fini"ti' . b_
of an mverse,
· ·
suppose a e G has two different inverses b and c then by
e ona• -eanda•c=e.
Then a * b = a * c
By using the left cancellation law,
b=c.
Hence the inverse of each element in a group is unique.
Corollary: Let G be a group. For all a, be G, we have (a* b)i =bi* ai.
Proof: Since G be a group·, for a, b e G there are uruque · e1ements a•,· b.1 e G such that a * ai = e· = ai * a ~
and b * bi = bi * b = e.
Now, (a* b) *(bi* ai) =a* (b *bi)* ai [using associative properly]
=(a* e) * ai
= a•ai
:::;::e ... (i) . t.

Also; (a* b) *(a* b)i .= e ... (ii)-

~y using (i) ~d (ii)
(a* b) * (a* b)i = (a* b) * (bi~ ai)
•
or (a* b)i =bi* ai (using left cancellation law) /

Thus, bi* ai is the unique inverse of a,r1b i.e. (a* b)i =bi* ai.
J
J
Finite Groups and Group Tables
Let G be a group with finite number of elements with binary operation. Taking th~ ·set of single
element that satisfies all properties of group. Definitely the element must be e (Identity element
of G) i.e. G = {e}. The only possible binary operation* on G is defined by e * e· = e. The identity
element is always its own inverse in every group.
Let us consider a group structure on a set of two elements. Since _o ne of the element must play the
role of identity so we can write G = {e~ a}: Now, if we try to find a table for a binary operation* on
{e, a} that gives a group structure on {e, a}. When giving a table for a group operation, we_shall
· always list the identity first, as in the following table. · .
* e a

e e a

a a ?
Since e is to be identity, so e * x = x * e = x.
 - Group and Subgroups w jcs§,ai 91
_11 x e {e, a}. We are forced to fill in the table as follows if* · to · ·
fot av , IS gtve a group. , . .
. e for each element a e G there must be unique m·verse i G ch ..1.. : •
5i11c • • . a e su mat a * a• = a• * a = e. Which
1
-
force that a = a itself (s mce a• - e obviousty
1 does not work). Hence,
* e a
e e a
a •
a e
[)efinition: If G is a group, then the order of grou~ is the number of elements in G and is denoted by IGI
Example .ffl: G '= {1, ·-1, 1, -i} is a group of ol der 4:
G = {l, w, w 2} is a group of order 3
G = {O} is a group of order 1
G = {a * b =ab, where a, b, e I}
Where, G is a group of order infinite.
. ,

Example 41: Let G = [ (: . !} a{ b, c, de R} Then p~ove that G is a group with respect addition
of matrices

-
Oosure: For A, B e G, where A = [~ ~]

B =[; !]; ~~A+ B =[:•:;: .::::J ~ G. .

1
1

Since, sum of two real numbers is' again real. ·)

Associative: For any A, B, C e G such that
A + (B + q = (A + B) + C [Matrix addition is always commutative]

Existe~ce of ad(Qtiops Identity: For fill:Y _A e ,9 there exi_sts O= [~ ~] e G such that A + 0 = 0 ,+ A =

A. i.e..A + O = c [a dbJ- . ~ [O:O' 07 = [ac ++ o,

oJ O ·db++ OJ
o [a b]d ~ A
= c
Similarly O+ A = A ·
0

Exi~,~ce ~f add,itive invers~: For any A'=[: .~] -~.Gthere exis~ .- A= [: ~ .=~] e G such that
a+ (-a) b+ (-b)] OJ
A+ (-A)= [ c + (-c) d + (-d) - 0 0
-[o
Similarly, -A + A = 0.
Thus .G is a group.

Example 42: G = [cca db)·· a, b, c, e R] Prove that

d G ·is not a Igroup with binary operation
, . . ,
multiplication of matrices.
lution: ' ~.
ForanyA,Be G,A=[: !J_a=[:: -~] , .. J· ·'., ., . .

CJ · [a b][~t
bt] [aa1 + bc1. abt + bd1l e
OSure: AB = c d Ct d1 = ca1 + dc1 , cb1 + d~J ~ . , ,.. ,
G
Since, aa1+ bc1, abt + bd1, ca1 + dc1, cbi + dd1e R . ·
 --
IIATBBIIATICS-U

Associative: Since matrix multiplication is associative i.e.

(AB) C = A(BC) for all A, B, C e G

Existence of identity: For any Ae G there exists I = [t ~] e G. Such that AI = IA = A.

Existence of inverse: For some matrix whose determinant is zero, we can't find A-1 such that
AA- 1 =I= A-1 A. Thus, G is not a group with binary operation multiplication. Here ( ; ; ) e
G but it's inverse not exist)

Example 43: Let G = [: :J, ad - ~c * 0, a, b, c, d, e RJ. Then.G is a group with binary operation
multiplication of matrices.

i. Oosure: For A = [: :J, B = [:: ::] e G, then ad - be* 0, a1d1 - btc1 * 0.

Now, I AB I = I A I I BI = (ad - be) (a1d1 - btc1) 0 *
AB is also invertible and AB e G.
ii. Matrix multiplication is assocjative i.e. VA, B, <;::: e G, we have (AB)C = A(BC).
iii.
·. . '[1 OJ
Existence of Identity: For any A e G there is identity matrix I = O l . Such tha~ AI = IA = A.

iv. Existence of multiplicative inverse: For any A e G there exists A-1 =ad-be -c 1 [d -bJ
a eG
Such that AA-1 =I= A-t A.
Thus, G is a group.
Example 44: Prove that C = (x + iy, x, y e RJ is· a gr~up -U1'der addi~on of complex number but is
not group under multiplication of complex numbers. t

Solution:
For multiplication (i) Let z, w e O where z = x + iy, w = x1 + iy1-
Then zw = (xx1 - yy1) + i (xy1 + yx1) e C. .
ii. Since we know that multiplication of complex numbers is associative i.e. z, w,'ue C. Then,
(zw)u = z (wu) ..
iii. Existence of multiplicative identity. For any z e C there exists 1 e ·C such that
z.1 = 1.z = z
iv. For O = O + i O e IC. We can't find the multiplicative inverse of Oin IC. Hence C is not group under
1
· multiplication of complex numbers. -' ·' '
Again, for addition binary operation, J

(i) z, w e C then z + w = (x + x1) + i (y + y1) e C

(ii) Addition of complex numbers is always associate
(iii) For any z = x + iy e C. There is O= 0 + i Oe C such that
z+O=z=O+z
(iv) For any z =x + iy,: C there exists ...
- z = -x - iy e C such that
z + (-z) =0 = (-z) + z
Thus, C is a group under addition. ·
Example 45: Check whether the following ~e ~o~p or no!. .
(a) G = {1, -1, i, -i} with multipbcation of comple~.number. ·
(b) G = {1, w, w2} with multiplicati~n ~f c~mplex numb~r, w_is an imaginary cube
root of unity. _
(c) z6 = {O, 1, 2,3, 4, 5} under the·multiplication. ·
(d) Z4 = {O, 1, 2, 3} under addition.
 Gn>up and hbpoape

ornol
1 l -1 i -i
-1 -1 -1 -i
i 1 -i i
i
E -i . -i -1 1
-1 .
1
From the table, we can obse th 1 -1
1. is the identj:ty element
respectively.
0
7'~ ·
~lemen~ ~f G _are closed under multiplication. It is clear that
e multipli~ative inverse of 1, -1, i, -i are 1, -1, -i and i
ll
It is obvious that element of G ar . .
(b) G = {l, w, w2} e associative under multiplication and hence G is a group.
,It will be easv by using multi Ii . b· r
tp cation ta le to.check whether G is l!IOUP or not
X 1 . w2
w
1 1 w w2
w w w2 1
w2 w2 1 w
From the table~ 1t IS c~ear ~at elements of G are closed under multiplication. From the table, we
can observe 1 IS the identity element of G. The multiplicative· inverse of 1 2 1 2
. . I . b . tha th l f , w, w are ' w ' w
respective y. It 1s o v1ous t e e ement o G satisfy the associative.
Thus, G is a group under multiplication.
i6 = (0, 1, 2, 3, 4, 5}
Themul tip · Iication ta ble for theelements ;z6
'x'6 o· ·. ·. 1 2 3 4 5
-0 0 0 0 , ' 0 0 0
1 0 1 2 - 3 5 4
2 0 2 4 0 4 2
'
3 0 3 o_ 3 3 0
·4 0 4 2 0 2 4
5 0 5 4 3 1 2
From the table, It is clear that elements of ikare·closed under multiplication. The number 1 play
the role of multiplicative identity. The numbers 0, 2;,3,,4 does n~t have their inverse and hence
Z6 is not group under multiplication.
d. A ain, For Z4 = {O, 1, 2, 3, }·under alidition
1 2 3
+4 0
0 O 1 2 3
1 1 2 3 0
2 2 ~ ~ ;
3 3 · I z are closed under addition. The number '0' is
0 4
From the table: !J is de~ _tha~ elemen~O o,
are 3, 2, 1 respectively.
123 1
additive identity. The additive inverse O ' ' ' ,
For associative: Let a= o,·b = l, c = 2 . _
3
([a] +4 [bl) +4 [c] = ([OJ +4 [11) +4 [2] = [1] +4 3[[;j =
[a) +4 ([b] +4 [cl) = [O] +4 ([1] +4 [21) = [O] +4 -
Again, a'= 3, b = 2, c = 1 then _ l] + [1] = 2
([a] +4 [b]) +4 [c) = ([3] +4 [21) +4 [1] ~ [[3] +44 [3] = 2 ..
[a] +4 ([b] +4 [cl) = [3] +4 ([2] +4 [l]) -
'thus, Z4is a group under addition.
 IIATBBIIATIC&-11

;---;~E--.-::Xl--=~R
7
(-::;-~I~S:-;--E-;-9-;;--.:7l---:-----=-------:-----=--=--------__J'
1. Determine whether the ~efini!i,on of * does give a binary operation on the set.
a. On Z, define * by letting a * b = 3a + 2b b. On z+, define * by letting a * b = ab
+ a+b
! , c. On I , define * by letting a * b = a - b d. On Z, define * by letting a * b =- 2 -
2. Prove that multiplication on a set G = {l, ro,ro2}, where ro is a' complex cube root of unity is a
binary operation on G.
. a~b
3. Determine the binary operation* defined on Q by letting a * b =- 2- is associative and commutative.
4. What three things must we check to-determine whether· a function cl> : S ➔ S' is an isomorphism
of a binary structure?
5. Test which of the following are isomorphic binary structure?
a. <Z, +> with <Z, +> where cl>(n) = -n for n e Z.
b. <Z, +> with <Z, +> where cl>(x) = 2x for X E z.
c. <Z, +> with <Z, +> where·«n) = n "+> i for n e Z.
. .
. ·x . .
<Q +> with <Q +>. where «x) == 2 for x e Q.
e.
d.
<Q · > with <Q · >where ci,(x) = x2 for x e Q. '· · 1.
..
'-

f. <Ii, · > with <Ii, · > where cl>(x) = x 3 for x e I. I ·~ ' ' •
g. <M2(R),- · > with <I, · >--_wherei(A) is the -determinant of matrix A.
-h. <M1 (R), ::> with (R, ~) ~here ~A) is ~e determinant of matrix A.
i. <I,+> ~th <ii+, . > where cp(r)1= (½} forr E: I I

Determine whether the b ~ operation * gives ~ group structure on the giv~n set-. - -
6.
a. Let • be defined on Z by letting a * b =. ab. . \.I
b. Let * be defined on 2 Z = {2n I n-e Z} by letting a * b = a-+ b.
c. Let * be defined on 1+ by letting-a -. b_.= ~ . • ·i, ! ,··

d. Let • be defined-on Q by letting a·-• b = ab. •:

~ ,1 I ,~ f a

t * be defined on the set 1+ of non-zero recµ numbers by letting a•b '? b .

e. Le
f. Let * be defined on C by letting a * b = I ab I ,
g. Let z be set of integers and * is the binary operation defined on Z by a "" b-= a + b + 1.

7.
. .
Let * be defined on Q+by a * b = 2ab .Show that (Q+, *) 1s• an abellan group.

(b) yes . (c) No (d) No

1. (a) Yes 4. One to one, onto and (\>(x * y) = (\>(x) * (\>(y) v- x, Y e S
3. commutative but not associative (b) No . (c) No (d) yes
5. (a) yes (f) yes (g) No, (h) No
(e) No
(i) Yes (b) Yes
6. (a) No, i.e. inverse of each element may not be in Z. No, i.e. inverse of Ois not defined
(d)
(c) No, i.e. elements are not associative No, i.e., there is no identity element
(f)
(e) No, i.e. elements are not associative.
(g) Yes
1
Since, <Z, +> is subgroup of <R
, +>but <Q• ·> · t b
Q• c R. Every G has subgroups G 't If , is no a su group of <Ii,+>, even though as sets
a 1 se .

EJSJJlple 1: G = {1, w, w2}, whete w is a cube r . .

then.H = {1} is a subgroup of G. oot of uruty is a group with respect to multiplication

Example 2: Let G = {1, -1 i -i} is a 'th .

G and H = {~l , l} . al group w1 respect to multiplication then H1 = {1} is ~ubgroup of
. ~ , is s~ subgroup of G. · .
Example 3: Let G = i be a grou :th ·.b. · - · · · ' · · - · · ··
P wi ~ mary operation addition then a subset H = {2n : n e iZ} is a
sub grou ofG. ·

Theorem: Let (G, *)'be a ioup ·then a'nort~mpty·subset"l-I °8£G is a ·subgroup of G if

(a) For all a, be H then a* be H (closure)

.,.
n'(t,j
'l .
~ti~-hn a+'e H theh ~i ·e :.H
.
~! - t ••
(E~isterice o/iri~erse}"
! , ·' ~ · ., · ~-. ,.. . ·1 ~... . .
·
' . ,
'·
•
Remark: From this theorem we can observe that for subgroµp it is suf#c;ient to test closure and
existence of inverse. ·
Sub-group Example

E~ample 4: G ., {(; , !): a, b, c;, _de R} is a.group,under additi.on, show that H = {(~ :): b, c, de R1
is a subgroup of G under addition.1 • • •

..
~Jgtjon:·. '
(0 O') . ' •. u.
The matrix 0 = \0 O) e G is also a member of H since for the membership of H the first entry
must be zero and rest entry are free. Thus, the identity ~lemei:t~ ~~ 0 of G is also identity element
of H. Hence H is non-empty subset of G.
(o bt, - (o· ~, + =( o ~ + bi\ e H.
Oosure: Let A, B e H then A = \Ct dJ , B - Ci · dJ tben A B \Ct + c2 d1 + dJ
- . - [ 0 -bil
0 bt . trix A- -dJ e H ~uch that
Existence of Inverse: For any AeH, A= [ct •dJ there 1S a ma - - · -Cl

A+(-A)=( 0
- Ct+ C2
-dbt++dbi\
- t J
J~~ ~) =(-A)+A
Thus, H is a subgroup under matrix ad~tion.
 MATHEMATICB-11 I ..

Example S: Let G = { (: :): ad - be ¢ 0} be a group under matrix multiplication, then show that

H = {(~ ~: ad¢0} is a subgroup of G.

Solution:

The id~ntity element I =[~ ~] e G is also identity element of H.

i.e. [~ ~] e H so H is non-empty subset of G.

Closure: For any A=[~ ~], B = [~

1
~J e H, AB=[~ . ~][~
1
1J 1
=[a; d~J
, '

Here aq ¢ 0, a1 d1 ¢ 0 then aa1.dd1 = ad a1d 1 ¢ O.

:. ABe H . .. . ..

En•::~~ ~m[T7] ;.!.~ :on sm~

l[n~]:

ad 0 . a 1
matrix A E

e H. Smee, a·d - ad·*0.

then A-1 exists an~ , I rn

0 - '
. a .
Thus, H is a subgroup of G.
Evmple 6: Let G = .fl, -1, i -i} is a group under _m ultiplicatio~ then ~ = {1, -1) is a sub group,of
' .
G- . ·l 1
Solution:
- ' , • ( 1 -JI

The identity element 1 of G is also in H . It is· clear that elements of H are closed
-.. ~• .1 • ' ). ·t
I ~•
under
i , f

multiplication. The inverse of 1 is 1 itself. Similarly the inverse of -1 is also -1 itself. This type of
i •

group is called comn\utative or abelian.' ' . t

_, .'l l I J

Example 7: Let G = {l, w, w 2} is a group then H = {1} is subgroup of G.

Solution: I

. -Here the !dentity element ofG is-also in H.·The in"\terse of 1 is,itself 1. q ostµe is 'Obvious. 'ffius,
H .i s subgroup. This type of subgroup is called triyial.subgrq~p 0 £ G. , r tjll ,
Example 8: Let z, = {O, 1, 2, 3) is a group under addition. Determine whether H = {0; ·1 ~10} is
subgroup ~f Z, or not. r _. , It •

Solution:
• l ' • ' (1

The addition modulo 4 fable for H is;

+, 0 1 3
0 0 1 3
1 1 2 0
3 3 0 2 r

Here, H is not closed under +, because 2 ~ H. Thus His not subgroup of L.

Generator of a group
If we consider Z6 = {O, 1, 2, 3, 4, 5} then we can observe that how large a su~gr~up ~ of Z6 ~hi~
would contains 2. It must have identity element Oand 2 ·+ 2 ~ 4. ·
.. ,H = {O, 2, 4} is a subgroup of Z6 under addition, and H = {2n: n e z+} = <2>. Here 2 is called
generator.
 Ormap and Subp,up. m IC&vrat9f
similarly for Z12, a subgroup which contains 3 . H -
every elements of H can be obtained b . IS - {0,3, 6, 9} and 3 is called generator. Since
y using 3 under the addition operation.
H ={3n, n e z•J =<3>.

Theorem: Let G be a group and let a e G. Then H = {a" I n e Z} is a subgroup of G and is the smallest
subgroup of G that contains a, that IS,
. every subgroup containing a contains H. •

\ .1_HXERf~ISE 9,"2
Determine whether the si".:en subset of the complex n~bers is .a subgroup of tl_te group C of
complex numbers under addition.
1. I
2. Q• l•

!er 3. 7Z = {O, 7, 14, .....} u

J
{- 7,,-:-14, .. -..I }l _J._ .•,1(...,
~ ,
. . , ,:, .\
,- ..
of 4. The set i I of pure imaginary numbers including 0.
5. The set JtQ of ra~onal ~ultiples of 1t.
6. The set G -= {Jtri. I n e Z}
From (7-11) Determine whether the given set-,of invertible n x n· matrices with real numbers
n. .
entries is a subgroup of GL (n, I) [group of n .x in~ertible matrices of real number]
7. The nxn matrices with determinant 2.
; 8. The diagonal nxn matrices with no zeros on the diagonal.
9. The upper-triangular nxn matrices with no zeros on the diagonal.
10. The nxn matrices with determinant -1.
11. H = (A: IA I is - 1 or 1}
~
Let p be the set of all real valued functions with domain I and let Fbe the subset of F consisting of,
those functions that have a non-zero value at every point in I. In a question no. 12 to 17; determine
whether the given subset of F with the induced operation is (a) a subgroup of group F under

addition, (b) a subgroup of the group Funder multiplication.

~
F
The subset of all f e F, such that f(l) = 0.

The subset of all f e Fsuch that f(l) = 1

15. ff• ff e F: f(O) = 1}
 --
MATHEMATIC&-D

~ ~
16. · The subset of all fe F such that f(O) = -1
17. The subset of all constant function in F.
18. Which of the following groups are cyclic? For each cyclic group, list all the generators of the
group. Gi = <Z, + >, G2 = <Q +>, ~ = <Q+, · >, G4 = <6~, +>
19. Find the order of the cyclic subgroup of Z4 generated by 3.

1. yes 2.
No 3. Yes 4. Yes
5. Yes -. 6. No 7. No
8. No under addition, yes under multiplication 9. - yes 10. No
11 . . Yes

12. (a) No, F is not closed under addition (b) Yes

13. (a) yes (b) _ No, it is not even subset of F

·:i ( •
14. (a) No; It is not dosed under aadition . (b) Yes
15. (a) No, it is not closed under addition (b) Yes
16. (a) No, it is not dosed under addition
(b) No, it is not closed under multiplication

(b) No, the zero constant function is not in F

~
17. (a) Yes
18. Gi = <Z,+>, -1 and are generator
19. 3

i ._ •• ri.
□□□
... r"

, I i

' t ,. i: ,,
 I • •.

•L •
RING AND FIELD
We , h . · .~ t)'frf,e ! -~!th a , singl' b!n_,a ry
·o perati ., 'et"'wittf ·two binary operations
, ~ ~ .'. ~,.: J • ~

addition· an 'OP,eratic>ns cal.l ed a ring when

'i-Y. -,,_
~ t
'.ff satisfy spin " , .we sh9lf try t~ develop furth~r
pr~p~;rtJeS" of rin,;, ., ' .. ·" ~-- w ,
nfa{~·:'cihci fiJld. " , · . · .
, ~.
· ..

LEARNING OUTCOMES

......, bapter you should be able to:

After studying ww, c
I I ,
& Ring and Basic Properties
., ' • , I

& Properties of Ring

Field
 MATRBMATICS-ll

an• Baal~ Propertiea

set R together with two binary op-,a

<R, +, .> is .if the following conditions are satisfied: ...,_ _. ·-••=
Rt: < R,, +> is abeliangroup.
under mullf iii.cation: for all a, b e R, ab e R
tiplication: for all a, b e R, (ab)c = a(bc). 2.
b, c e R, a(b + c) •ab+ ac (left distributive)
a, b, c e R, (a + b)c = ac + be (right distributive)
3.
DeRnltlon:A ring <"R +, .> is called .commutative ring if it is commutative under ·
operation multiplication i.e. for all a, b e R, ab = ba.
A set A having a single element O with two binary operations t}ddition and multiplication ·
• ,)&;, •

called-a zero ring · · · "'!:,· ., '· · .

• • : > ·~:~\ • ~=~ ' ·~ ' ~ , ~~ ' '
A ring R is ring with identity if it has.ml,l].tip}i.cati\7_·· n~ty element.i.e. for every a e R th
existle Rs.la.1=1.a=a·- · .=· : · ' , ; . ; \ ' {tt•"·· ·
s'$\ ·), ,
4
Example 1: The set of all integers Z is commutative ring under binary operation addition and
multiplication i.e. <Z, +, ·> is commutative ring.
Example 2: Also <Q +, .>, <R, +, .<, <C, +,.>are rings. All these ring are commutative ring and ring
with identity.
Example ·3: The set of all integers Z· is commutative ring under tiinary operation addition and
multiplication. · ;·
Solution:
We have, z = {... , - 3, - 2,.:.. 1,0; 1, 2, 3, ...-}
1. For abelian group under addition
i. closure: "I a, b eZ we have a+ be Z
e.g. it a = - 3, b = 2 then a + b = - 1 e Z
ii. Associative: "I a, b,'c e Z we have (a + b) + c = a+ (b + c)
e.g. if a = - 2, b = 3, c = 1 then
(a+ b) + c =(- 2 + 3) + 1 =1 + 1 = 2
a + (b + C) = - 2 + (~ + 1) ~ - 2 + ·4 = 2

So it is associative
iii. Existence of identity
Since there is an element 0 e Z s. t.
0+a=a+0=a'v'ae Z
iv. Existence of inverse

'v'a e Z there exist-a e Z such that

 Rlnp and Field Ill ~
a + ( - a)= (-a)+ a~= 0
e.g. if 2, 3, ... are in Z th I ,
13
en-2 3
v. Commutative ' - ' · · · also are in z
y
z we have a + b - b
"i/ a, b, e
. - +a
e.g. if a = - 3, b = 2 then
a+b=-3+2=-1 ;:i

2. a osed under multi Ii .

b+a=2-3=
-1
P cation
"i/ a, b ~z we have ab e z
eg. if a = - 3, b = 4 th
3. As . . enab=(-3) ·4=-12eZ
SOC1ative under multiPIication
.
'ti a, b, c e Z we have a (b ) _ (
. · · c - a. b). c
e.g. if a = 2, b = - 3, c = 5 .
then a. (b. c) = 2. (- 3. 5) = 2. (-15) = - 30
(a · b) • c= (2. - 3) . 5 ·
=-6. 5 f1
=-30
Distributive law
~ b .
a, , c, E Z we have a . (b + c) = a . b + a . c and
(a + b) • c = a . c + b . c
eg. if a = S , b = 2, c = - 3 we have
&. • • • ,_

a . ~ + c) = S . (2 - 3) = 5 . (-1) = - 5
a. b +a. c = S. 2 + 5.(-3) = 10-15 = -5
S~ly,
= ( S + 2). (- 3) = 7. (- 3) = - 21
(a+ b). c
a. c + b. c = 5. (-3) + 2. (-3) = -15-6 =·-21 1-
Hence, (z, +,.)forms a ring. ! I

Example 4: The set of all nxn mattjces Mn(R) whose entri~s _are real numbers is ring under biruµy
operation addition and multiplication i.e. <M2(R), +,.>is ~g, but it is not commutative
ring. In particular, ·

M(R) = {~::
2 :::} aq e R} is ring and is ring with identity, because (t
multiplicative identity. But M2(R) is not commutative. Because

G~ a -1)
-D=C: 2 and
a-DG V=C!i
G Da
 MATHEMATICS-ll ·

Example 5: The set <22, +> is commutative ring but not ring with identity because 1 e -rz.
Example 6: Is set (a + ln/2. :
a, b e Z} with usual addition and multiplication binary operation is
ring? Whether commutative ring or not.
Solution:
Let, R = (a + W2, : a, b e Z} and binary operation defined as
(a+ W2,) + (c + bV2-) = (a+ c) + (b + d) V2 ......(i)
and (a + W2,) (c + dV2) = (ac + 2bd) + (ad + be) V2 .......(ii)
For Rl: We show R is abelian group under binary operation addition i.e. we show <R, +> is abelian
group.
Gl: R is closed under binary operation addition.
Addition binary operation defined as in (i), says that R is closed.
G2: R is associative under binary operation addition.
Let x = a +b\/2 ,y =c + dV2 and z = e + f-.{2. e R, so a, b, c, d, e, £e Z
Here, x + (y + z) = (a+ b\/2) + [(c + e) + (d + £) V2]
=[a+ c + e] + (b + d + f)V2
Similarly we find (x + y) + z = (a + c + e) + (b + d + £) V2
Thus (x + y) + z = x + (y + z), hence R is associative.
G3: Since O e Z so; 0 = 0 + Ch/2 e R as Oe Z is the additati:ye identity element of R, because for every
x=a+lrJ2, e R;
0 + x = 0 + o{2 + a + lY\/2 = a + l1\J2 = x,
Hence O = 0 + o{2 act as identity element in R.
G4: For every a + lY\[2. e R so a, b e Z hence - a, - b·e Z thus
- a + (-b) "2, e R such that (a + W2) + (-a+ (-b) -[2.) = 0
Hence for every a+ lY\[2. e R · 3 inverse element- a -+·(~b)-{2. e R. · ,,,
I 1
GS: For every x = a + b\J2 , y = c + d"2, e R:·where a, b, c, d e z.'
x+y =· (a + b\/2.) + (c + d"2,)

= (a + c) + (b + d) "2,
. ll.

= (c + a) + (d + b) -[2.

= (c + d-{2,) +_(a + b\/2.)

=y+x
Thus <R, +> is abelian group.
For R2: R is closed under binary operation multiplication
Multiplication binary operator defined as in (ii) says that R is closed.
 for R3: Associative under Illulti Ii· .
,• P cation• Let r; i...
Rlup ..... l'leJd U. §'.,.,.
__ _.
m
b,c,d,e,fe' Z . · . • _ · "~a'll' uy2 Y~c+""'i
>n is , •• ~ ' ,., ~ V"- '-Ii e Rr W here a
and z ==· e + '-V~
Then (xy) z = [(ac + 2bd) + (ad be) .. r;.2
• I
:! ,1 I •
,

_ V"-]+(e+f\[2.) . , '·
- (ac + -2bd):e + 2( d
_ - .. '!.. "' !ic)f + [(ac + ~bd) f +(ad+ be) e)...[2
- (ace + 2bde) + 2(adf + be . l '

S:lllwar
-: 1 1y, we get •· f) + [acf :t, 2bdf +(ade + bee)] ..v~
0_
f JI •

x(yz)
= (a+ b\/2) [(ce + 2df) + (cf+ de) V2 J
- [ace+ 2adf ·
- + (bcf + bde)2) + (acf + ade) +bee+ 2bdfJ V2
{j{J > 1 r I

= (ace + 2bde) + 2(adf + bcf)J + [ad+ ad; + 2bdf + bee) {2

= (xy)z .\

For R4: Distributive: Here (x + y)z =[(a+ c) + (b + dh{2] (e + fl/2)

= (ae + ce + 2bf +~elf)+ [af +cf+ be+ deh/2

and xz + yz = (ae + 2bf) + (af + be) "2, + (ce + 2df) + (cf+ de) V2

.,
'•
. .
Similarly we s1-low-"x(y *-z) = xy +,xz.
• ( l. ' ,.

H ~ thesetR= [a, + 'rrfi; '~, ·b'.e 'ZJ ~rih~

For commutative ring,

Here, x . y = (a + bV2) (c +-d-{2 )= (~c·+ 2bd) + (ad+ be) \[2.

and, y . x = (c + cJV2) (a + b\[2.) = (ca + 2db) + (ch + da) V2
= (ac + 2bd) + (be + ad) V2 ' ll

= (ac + 2bd) +(ad+'~ ) "{2;·. {f!I ~: .! .;, ~- r•

fl
Sincex.y=y.x I

So it is commutative ring. . der bina,y oper,,tion additiop an4 multiplication,

E Pie 7: Set Z1 = fO, 1, 2, 3, 4, 5, 6} form nng un , . - - .
lllllllmodulo 7. 1 • .~ • 1 ' r ' ,,
The composition ta bles are •7 ·O 1 2 3 4 5 6
1 2 3 4 5 6 I
+7 0 0 0 0 0 0 0
4 5 6 0 0
0 0 1 2 3 1 . s·
6 0 1 0 . 2 3 4 6
I"

1 1 2 3 4 5 1 3 5,
1 2 0 2 4 6
6 0
3
2
3
2 3
4
4
5
5
6 .. 0 1I
.2 , .. 3, 0 3
·o 4 1
6 > 2
.. 5
5
2
1 . 4
6 3
2 3 4 2,
4 4 5 6 0 1 .o 5 3 1 6 4
3 4 5 1
5 5 6 0 1 2 0 6 5 4 I 3 L
2
5 6
4 -
• £

6 6 0 1 2 3 ~

Table 2
Table 1 - -
 MATHEMATICS-II

Rl: We show <Z1, +> is abelian group. From composition Table (1). Z1 is closed. Also it is assoctate.
Here O act as additive identity and every elements has additive inverse. Inverse of Ois itseli. Inverse
of 1 is 6. Inverse of 2 is 5 and inverse of 3 is 4. Also 2 7 is abelian group under binary operation
addition.
R2: From composition Table (2), 2 7 is closed under binary operation·muJ.tiplication.
R3: Using composition Table (2), 2 7 is associative under multiplication.
For example: 2, 4, 5 e 2 7 and
2.(4 . 5) = 2(6) = 5 and
(2. 4). 5 =1.5 =5
R4: Distributive: Using composition table (1) and (2), 2 7 holds both distributive laws.
For example: 3(4 + 6) = 3(3) = 2
and 3.4 + 3.6 = 5 + 4 = 2
3(4 + 6) = 3 . 4 +·3. 6
Similarly we show right distributive.
Hence, Z1 is ring.

Example 8: Let <R, +,.>and <S, +,.>are two rings then RxS =· {(r, s): for .all re R, s ·e S} is set of all
ordered 2-tuple form ring denoted <RxS, +,.>,where addition and multiplication binary
operators are defined as (r1, s1) + (r2, s2) = (r1 + r2, s1 + s2) and (r1, s1) . (r2, s2) ,=; (r1r2, s1S2)._
Here additive identity of RxS is (0, 0) and multiplicative identin7 of_~xS is (1, 1).
Additive inverse of (r, s) = (r, s)i ·
= (ri, si)
. ·.,
Where ri is additive inverse of r e R I I f -
I. •/Y 1 . / '

si is additive inverse of s e S
Example 9: Compute the product in 'giva:1- r~gs:
(i) (11) (-4) in Z1s (ii) (2, 3) (3, 5) E Zs X z9 r •I

Solution
(i) · In Z1s, -4 ='11, 50 (ll) (-4) =(1'1) (ll) = l. ...
(ii) (2, 3). (3, 5) = (2 Xs3, 3 X95) = (1, 6)
Example ·to: If (2, 3), (3, 5) e Zs-x 29.'F~d (2, 3) . (3, 5) and (2, 3) + (3, ~)

Here (2, 3) + (3, 5) =(2 +s3, 3+95) =.(O, 8)

• I ,. • f""

and (2, 3) (3, 5) = (2 x5 3, 3 x95) =(1, 6)

• d th dd"ti" e m·verse and multiplicative inverse of element (2, 3) in Z3 x Zs. ·
Example 11:Fm ~ a 1 v .
t
For additive inverse of (2, 3)

Here, 2 e ZJ and 3 e Zs
Additive inverse of 2 e 23 is 1
 ,4dditive inverse of 3 e Zs is 2 - . . 11114 Pleld
~ l§#ri- 1!
e.
e 11tUS additive inverse of (2, 3) is (l ) . ,
, 2 • Beca\lse (2 1
I •
por multiplicative inverse of (2, 3) ' 3) + (1, 2) = = (2 +
• ' I ~1 , l 1' 3 +5 2) a (0, Q)
I'
,Jultiplicative inverse of 2 e Za is 2 ·

,Jultiplicative inverse of 3 e Zs is
2
'fhus multiplicative inverse of (2, 3) is ( ' , . ,
2, 2). BecaUSe (2 3)
Example 12:Check set 2Z>cZ with b" . ' · (2, 2) = (2 ><32, 3 ><s2) = (l, l). . .
· mary
operation additi
nng or not? H it is rino then ta On ,and multiplication J... co . .
---o .. s te w};\etha co .. .,, mponents IS
Given 2Z>cZ = {(2m, n) : for all,.,,,. Z} '· · -~utaqve, wh~ther it has ~ity?
.... ., n e and additi ·
For (2m1, n1), ( ~ Ill) e 2Z >c Z . on ~d Jllultiplication is defin~ ~
ti
(2m1, n1) + ( ~ Ill) = (2(m1 + ~2), n1 + 112)
...... (i) r; • I; •
(2m1, n1) . (2m2, Ill) = (2(2P"t1 Dl.2), n1112) .,
·:• ..•.(ii) , fr!
Rl: (22 x Z, +> is ~belian group.

Binary o~r~tion d~ed in (i) sh~w~ ~t 2Z x Z is clo~d un~~ binary o~ ration addition. For
X = (2m1, n1), Y =~~Ill) ~d Z = (~,: 1:13)"1~ 2?, X ,z ri .II ,. , l· • · t .~ ♦I
x + (y + z) ~ (2m1, n1) + [2(mi + ~), n2 + -~ )] = (2(m1 + m2 + Dl3), n1 + n2 + Il3) , •,.
and (x + y) + z = [2(m1 + m2), (n1 + n2)] + (2Dl3, fll) =.(2(m1 + m2 + Il3), n1-+ n2+ Il3)
Thus 2Z x Z is associative under binary operation addition.
Since (2 ><
.. --
----~· -
0, 0) = (0~ 0) e 2Z><Z act as additive identity element because for all (2m, n) e 2Z x Z,
·1 •

(2m, n) + (0, 0) = (2m + 0, n + 0) = (2m, n) and (0~ 0) + (2m, n) = (0 -t1'2m, 0 + n) = (2m, h)

For every (2.m, n) e 2Z >< z 3 (- 2m, -n) e 2Z><Z-, -·sudh that · " ·~
I,
(2m, n) + (~2m, _ n) = (0, 0) so (-2in, · n) is ·adaitive ifl\retse·of (2m, n).
, I f •l r ir r
Let X = (2m~ n1) and Y =(2m2, n2) .E 2Z><Z then 1, • ......

x + y = (2(m1 + m2), n1 + n2) = (2m1 + 2m2, n1 + n2) . • I. ,I I t ' , .,. .. . '

',

= (2m2 + 2m1, n2 + n1)

= (2(m2 + m1), niI,+,._,n1) ·.1 . ! .

=y+x /) I

,,7 Z is commutative. .
:. ~x · .. sh that Tl x z is closed under multiplicatio~tbinary operation.
tion defined on (n), ow
R2: BinarY opera . Let - (2m n1) y =(2m2,•ni) and z =(2ms; na) e 2Z ?< Z
der multiplication: x- 1, '
R3: ,Associative un =(4mim2, nin2). (2m3. na) =(8m1m21l\3, n1n21\3)
) (2JJl2, n2)](2n13, I13) · , I' ) I J •
(xy)z = [(2JJ11, n1 . ) (2n13 ru)] = (2m1, n1) .(~21113,nill3) = (8m11Il2ffil, n1mll3 ·. ,-tl
2JJl2, n2 '
x(yz) = (2JI11, n1) [(
:. (xy)z = x(yZ)
 C
MATHEMATICS-II

R4: Distributive: For left distributive

'?'(y + z) = (2mi, n1) . [2(rn.i + 11\J), n2 + Il3)
= (4m.1(m2 + 11\J), n1(n2 + Il3)) = (4m1m2 + 4m11I\J, n1n2 + n1Il3)
and xy + xz = (4m.1m2, n1n2) + (4m11I\J, n1Il3)
= (4m.1m2 + 4m.11I\J, n1n2 + n1Il3)
Hence x(y + z) = xy + xz ✓ I•

Similarly we show right distributive (x + y)z = xz + yz.

Thus 2ZxZ is ring under binary operation addition and multiplication. r•

This ring is commutative {abelian) also: because

xy = (4m.1m2, n1n2) and
yx = (4m2m1, n2n1) = (4m1m2, n1n2) = xy I •

It has no unity elements because 2z has no multiplicative identity element 1. ·

Proo{;. _ ,. - ._· r •

(1) We know that a ,. 0 = ;i. ·(O + ·o)

= a.O + a.O ..... (i) •(by distributive law)
Also, a.O = a.O + 0 ••.. (ii) (definitioI), of identity)_

From (i) and (ii), - .,, n n.'-

Hence a.O + a.O = a.O + 0

a.O = o (By left canc~llation law) I +- I ',
Since, 0.a = (0 + 0) • a . ' : .
=O.a+ 0 .a ..... (iii) (By distributive law)
Also O.a = 0 + O. a ...... (iv) (By def. of idehtity)

From (iii) and (iv)

o.a + o. a = o+ '6. a J~..l'

. ,. ,, . \

So, O.a = o By right cancellation law.

Hence, a . 0 = 0 . a = 0 ,.
I
(2) We know that ab"!" [- (ab)]-= 0, .... (i) (by def. of inxerse) .I I

From (1)
0 = a.O
 = a (b + (-b)J
..I I ~ J /' •

=ab+ a(-b) ,,,

..... (ii) (By dis . .
from (i) and {ii)' , . tributive law)
f. , I

r /I / I
ab+ a{-b) ==ab+ [
- (ab)J l
a(-b) == 7(ab) , J ,, .,
1j I
I
• f 'rqm
Also fro · · · (A) '
~ !1)
>
, ' (By left cancellati I
0 ,== 0.b ... . oi:i awl

=[a+ (~)lb
-- . ... ' , '

=ab+ (-a).b c·)

psmg (i) anq (iii) ....•. w (By distributive law) ·.

ab+ (-a)b =ab+ [- (ab)] :'

.. . (-a)b == i (ab) _ ,
Hence, from (A).~d ID.) . , - ... ,(B) (By-left cancellation l~w) .. -
.... (3) - Since (-a) (-=b) +
I .. : ,
\Y w~ have a (-J;,) ~ (
r.
-- - -
- ·
.
a J -b)_ = (ra_ +a) (-b)
, ·
a) P ~•..-:.(aQ) · ,
-
1
- - . -

- ,.
· = o (-b) , r · --- = ~ -~ _ , • _
1
I I f t
- !.- ~ i,
' =O ' , I
I
• I
'
-.L
=ao I

=a(b + (-b)) ' J- r,, f , ! t 1 -.,

= ab +a(-b)
... (.:a) (-b) + a (-b) = ab + a(-b)
(-a) (-b) =ab (By right cancellatipn law) -.. I • ✓

Unit Element: Let R be • th · ' · . ·-

a nng en element ae R is called unit eleme t if . has . .
In ring z has only two unit elem~nts whi~h are 1 and -1. n it multiplicative inverse.

In ring R all non-zero elements are unit elements. t' ' · ,.

In ring Zs, ~lement 1, 2, 3, 4 are the unit elements.

In ring Zs element 1, 3 are the unit elements.

A commutative ring with identity is called field if its every non-zero elements are unit element (unit
elements means non-zero elements which has multiplicative inverse). ,
Example 13: Set R and Q are ring under binary operptj~n addition ~ d multiplic~tion, and are also
field. Because they are commutative ring with identity ~d every non-zero elements
· has multiplicative inverse. . f.
 ◄
MATHEMATICS-D

Example 14: Z is not field, because only 1 and -1 has multiplicative inverse, other elemen'ls has no l
multiplicative inverse.
Example 1S: M2(R) is not field, there are so many elements has no multiplicative inverse, for

example, (! ;) e M2(R) but it has no multipli~ative inverse. ' ~-

Example 16: Ring <Zs, +, .>, is field but <Z6, +, · > is not field, because Zs = {O, l, 2, 3, 4} is
commutative ring with identity and also every non-zero elements are unit element
(inverse of 1 is 1, inverse of 2 is 3 and inverse of 4 is itself). In ring <Z6, +, .>, 1 and 5 are
unit elements but 2, 3 and 4 are not unit elements. Which we can see from table

"s 0 1 2 3 4
1
"6 0
. 1
I
·2 3' 4 5

0 0 0 0 0 0 0 0 0 0 o' ·o 0
I

1 0 1 2 3 4 5
0 1 2 3 4 1
2 4
2 0 2 4 1
'
3 2 0 ·2 4 .. 0
3 0 3 1 4 2 3 Q 3
..
0 · .~ 0 3
- ·O 4 2
4 0 4 3 2 1 4 0 4 2
I

5 0 5 4 3 2 1

11111 <.zp, +, .> is field, where p is prime. So Z2, 2 3, Zs, 2 7, 2 11, etc. are field.

Example 17:Solve the equation x2 + 2x + 4 = 0 in Z 6•

Solution: ·
When x = 0, x2 + 2x + 4 = 4:,:. 0 ..,, .. . t!' .

X = 1, x2 + 2x + 4 = 7 :I:- 0

X = 2, ,c2 + 2x + 4 •= 12 = 0.
.., ( . . .
X = 3, x2 + 2x + 4 = 19 :I:- 0

X = 4, x2 + 2x + 4 = 28 ':F- 0
·• · J
X = 5,_x2 + 2x + 4 = 39 ':F- 0

Solution of x2 + 2x + 4 =0 in Z6 is x =2.

Example 18:~ind all solutions-of equation x3 - 2x2 - 3x = O.in 2 12. -

Solution: 1•
•• I

. ' . ·1,
, '
Here,
J f ,I • ; 9 l ••
,. .
x 3- 2x 2- 3x = x {x 2- 2x - 3) = x(x + 1).(x - 3).
_,, .. • n

Thus x = 0, x = 3 are obvious solution-of x3 - 2x2 - 3x == 0 in Z12~ Not~ th~t ~x_== -1•~ ll is also solution
in Z12.
no for other solutions
Rlnp lllld J'feld CQ §fin
1~
x = 1; x3 - 2x2 - 3x = x(x + 1)
Or (x - 3) =-4 =8 * 0
x = 2; x3 - 2x2 - 3x = x(x + 1) (
. X - 3) == -6 == 6 * 0
X = 4· x3 - 2,c2 3
, - X= X(:x + 1) (
ls X-3) = 20*0
Lt
X = 5; x3 - 2,c2 3
- x = x(x + 1) (
x - 3) =5 >< 6 >< 2 =60 = 0·
x=6-x3 2x2 -3x - (
I - • • • ·
- X X + 1) (x - 3) = 6 >< 7 >< 3 = 126 ;,t 0
X = 7· x3 _ 2,c2
, - 3x = x(x + 1) (x - 3) = 7 >< 8 >< 4 = 224 ;,t 0
X= 8· x3 _2,c2 3
, - x = x(x + 1) (x - 3) = 8 >< 9 >< 5 = 360 = 0
X = 9; x3 _2x2 3 _
- x - x(x + 1) (x - 3) = 9 >< 10 >< 6 = 540 = 0
. ~ = 10; ,c3 - 2,c2 - 3x'= x(x + 1) (~ - 3) = 10 >< 11 >< 7'.;. 770 ;,t 0
X = 11 · ,c3 - 2,c2·- 3 - (
, x - xx+ 1) (x - 3) = 11 >< 12 >< 8 = 1056 = 0
,h ~ I .- , , '

x = 0, 3, 5, 8, 9 and 11 are also root of x3 - 2x2 - 3 = ,. . .

equation. x O. m 212. So, they are solutions of given . ,

Zero Divisor

ff a and bare two non-zero elements of ring R such that ab = othen a and bare called zero divisor.
Example 19: Find Zero divisor of ring z~. · . ,
( . ' .
Solution:
( I ..

Since Z12 = {O, 1, 2, 3, 4, S, 6, 7, 8, 9, 10, 11} and

-~\ ..... · ·~ n ' • 111

2.6 = 3.4 = 3.8 = 4.6 = 4.9 =6.6 = 6.8 =6.10 =8.9 =0.
Thus 2, 3, 4, 6, 8, 9 and 10 are the zero divisor of ring Z12. I\J -1 = 1
• ..

1111 The z.ero divisor are those elements in Zu whose gcd (greatest common~vik) witnll is not 1. l

Example 20: Find zero divisor of ring ~10-

Solution:

Here, .. ; 1 ,..._. 1·1 r .. ,, p •II" '1 Vflillr. 1 IW , , ,,. f

L, · ~ • · ,
I
:JO= {O, 1, 2, 3, 4, S, 6, 7, 8, 9}
f • ~ ' J f .- , ,

gcd (10, 2) = 2 * 1 gcd (10, 3) ;::: 1

I, gc:ci (W; 6} =2 :11: 1

gcd (10, 5) = 5 * 1

, gcd (10, By= 2* 1 gcd (ld, 9) ~ 1I · •" 'T f '1

..
,
. ,.
, ) r-

ore zero divisor of ring Z10 are 2, 4, S, .6, B. ·1 • · 1 ·:

 ◄
MATHEMATICS-D

Example 20: Find zero divisor of ring z 7•

Solution:
Here,
Z1 = (0, 1, 2, 3, 4, 5, 6}

gcd(7, 1) = 1
gcd (7, 2):: 1
gcd (7, 3) = 1
gcd(7, 4) = 1
gcd (7, 5) = 1
gcd (7, 6) = 1
There are no elements in 21 whose gcd 'th 7 .
WI IS not 1.
Hence Z1 has no zero divisor.

- Ring Zp, where P is P~e has no zero divt'sor.

• I _:. • .t

Example 21: Show that matrix[~

~] is a zero _divi5<?r o_f ring M 2(Z).

Let[xz YJ e M2(Z) so that

w

[~ ·;1c: ~]=[g gJ
1 -

x+2z=O .... (1)

' 2x +4z =o· ....(2) I • I

. y+2w=0 .... (3) I

-..----:,a
2y+4w=0 .... (4) I ..,,
--
►

From (1) and (2) when z =1, x =- 2

From (2) and (3) when w = 1, y = - 2

1 2J[-2
Thus, [ 2 4 1
-2J ro
t ~Lo
01
oJ. .
• ' , I • 11 1 I •

1 2] ·
Hence, [ 2 . 4 is a zero divisor in M2(Z). 1\- I l

Integral Domain
l
, A commutative ring are with identity having no zero divisor is called integral domain.
. .
F;or example: The ring (Q +, 1, (Z, +, · ), (R, +, ·) and (Zs,+, ·) are integral do~. Bu( ring Z~'2Z
and M2(Z) are not integral domain. . , ., . 1 • • , J

Example 22:Prove that ring Z10 is not integral domain. ..:. '< •
I •
.J '

Since Z10 = {O, 1, 2, 3, 4, 5, 6, 7, 8, 9J is ~g ~der binary ope~atiQn ,+1~ ~q 0 10. It is commuta~v~ ring
and is with identity also . .But 2 10 has zero divisor.' The zero divisor are 2, 4, 5, 6, 8, because
2.5 = 0, 4.5 =O; 5.6 =0, 8.5 =0. Hence Z10 is not integral domain. · ·
 - Every finite integral domain is field.
Example 23: By taking q,~e suitable example prove that every finite integral.domain is field.
Solution:
We know that Zs = {O, 1, 2, _3, 4} is ring under binary operation +s and 0
5. It is finite,
commutative and with identity. Also it has no zero divisor, because gcd (5, 1) =·1, gcd (5, 2) = 1,
gcd (5, 3) = 1, gcd (5, 4) = 1. Hence Zs is finite integral domain. To show Zs is'field, it is sufficient
to show every non-zero elements Qf, Zs has multiplicative inverse. Inverse of 1 is itself 1. Inverse
of 2 is 3 and inverse of 4 is itself.

/L.__:I_.. _Ex_,_~1,_(~_1s_~,~_1_(»_ _ _ _ _ _ _ _ ____,,

1. Compute the product in the given ring
(a) (12) (6) e Z25
(b) (20) (-8) e Z26
(c) (-3, 5) (2, -4) e 24 x Zu

2. Prove that given set with indicated operations are ring or not? If a ring is not f~rmed, t~ll why
· 1s
this is the case. If nng · formed state whether the ring is commutative, whether 1t has uruty and
whether is field.

(a) nZ with the usual addition and multiplication

l
Z x z with the usual addition and multiplication by components
(b)
(c) z+ with the usual addition and multiplication by components.
g .. d ulti lication by components.
(d) {a + b\j2. : a, b e Q} usual addition an m P
e . li ti by components.
(e) M2(R) with the usual addition and multip ca on
 IIATHBMATICS-D

3. Find the additive and" multiplicative inverse of the element (3, 2) in ring 4xZ7.
4. Solve the equation x2 - 5x + 6 = 0 in Zu.
5. Solve the equation 3x = 2 in 27.
6. Find the zero divisors of following rings.
(a) Z20 (b) Z16 (c) Zn

7. [2-lJ
Prove that 4 _2 is a zero divisor of ring M2(Z).

8. Prove that ring 2 7 is integral domain.

9. Prove that ring Z12 is not integral domain.
10. Prove that ring 2 11 is field.

1. (a) 22 (b) 22 , l (a) (2, 2) •, r'. I

3. (1, 5) ~d (3, 4) ~- x; ~~2, 3, 6, 11_ >I .., : 5. x=3 , .,.-1r :•r
6. (a) 2, 4, 5, 6, 8, 10, 12, 14, ,1s, 16, 1s I. .,
(b) 2, 4, 6, 8, 10, 12, 14 ,, I (c} has no zel"O'ruvisors ; '''J I) • i •• • I \ •;. )! •

I ) <

□□□

• >•

.,

f-;I
., I({, . I I
,, 1 .
'.: , J J ''
I ,'f .;

·1. ., ' J ""' fr..:.Ul , ., ' ........ iL

r ..
t i 'I b t ,-, I J

r_. I

. ·r
.
 Blbllography

BIBLIOGRAPH\'
l { T
, tu i ~ I
Anton: ~- ~ d Rorres, C:, Elementary Linear Algel,ra, John Wiley, 2011. ~ ' ~ u·
1~ 1
• ' 1

Bretscb er, 0., linear Algebra with Application, Prentice Hall, 20004: > ,

Datta, K.'B., ~~ and Line~r Alg~bra, Prentice Hall of India Pvt. Ltd., 2002.
1 1
> mr:-. l '.
Hoffman, K, and Kunze, R., Linear Algebra (Ilnd Edition), Pre;,_ticJe '.Hall of hldta Pvt. Ltd., 2002. ., .>
Jnawali, J., Gautam, R., and Sapkota, B.P ., A _
textbook ok algebra II, Sukunda Pustak Bhawan, Nepal,
2013. · ~ · .' . . . .
Jc;>hn ~- Fralei~, Victor J. Katz: A first course in abstract algebra. Narragansett, RI. '
#'- .. ) I ... 1) • t,. .. t • ( I # ( , ,. , • , • I •

Kolman,-B.., Elementary Linear Alge!»'q !'}ith :N11J_licati~, Pearspn,,2<>q7. J. • ~

Lax, D. Peter. (Ilnd Edition). Linear Algebra and Its Application. New York: Wiley Interscience A
John Wiley & Sons. Inc. 2007. ~
Lay, D.C., Linear Algebra and its applications (lllrd Edition), Pearson Education, (Indian edition),
2014. - ' . · i l ~-. -::;r_ ,€'- • _.,
,, ,r
Marwaha, A., An Introduction to Linear Algebra, Plil Leaming Pvt 'Ltd., 2014.
• ,~ . r • - ) , .. '· • • • (' / J •

Meyer, C. D., ·Matrix Analysis and Applied Linea.r Alg~br~, ~~, 2002.~ __ . . 0

Olver, P.J., Applied Linear Algebra, Pearson Education, 2005.

Polle, D., Linear Algebra, Cen~age Learning, 2011.
Robert, A.M., Linear Algebra: Example and Application, yvorld Scientific Publishing Company, 2005.
Sad~ L., Applied Linear Algebra (Ilnd Edition), American Mathe~tjcal ~iety, (lndi~ ':_~ition)

stan:~-Linear Algebra and ,tJ Appiication: eeng~ge-~g, :foos. ,, · · -~· · ·' J .,

l I

I ),.: _=: -= • I l f 1.
r • ., , !'

..... '

I .

....
• I. r
' . ,
I •
 ....
· MATBBMATICS-D

Model Question
Tribhuvan University
Institute of Science and Technology
Bachelor Level/ First Year/Second Semester/ Science Full Mark: 80
Computer Science and Technology (MfH 163) Pass Marks: 32
Mathematics II Time: 3 hrs.
Candidates are required to give their answers in their 0 ~ words as far as practica~le.
'
,,
Group A (10 x 3 = 30)
Attempt any THREE q~esti~ns. . ., · ,
1. What is pivot position? Apply elementary row operation to transform the following matrix
first into echelon form and then into reduced echelon form: . . .
o 3 6 4 -5] ·•
I
-6 i· · ""
3 -7 8
-5 8 9 '· l
3 J~ 12 ,;-~ , ~
r ,15
2. Define linear transformation with an example. Check the following transformation is linear
or not? T: !Jl.2 ➔ 9l2 be defined by , , , , ,, ,.
T(x, y) = (x, 2y).Also, let T(x, y) = (3x + y, 5x + 7y, x + 3y). Show that T is a one- to-one linear
transformation. Does T maps f/l,2 onto !Jl}? [3+ 2+5]
3. J.ind the LU factorization of : .. · r. ~

••
o 4 -1
s
-4 -5 •. ) 3 . ,-8 I
-2]
,1 '
•. J (

[2 -5 -4 1 . ,8
. --6 0 7 -3 ~ i1
4. Find a least square solution of the.inconsistent system Ax= b for ._

A=r~ b=rnJ n Group B (10 x 5 = 50)

Attempt any TEN questions.

5. Compute u + v, u-2v and 2u + vwhere u =[-}], v =[~]. [51

6. [2 ,' OJ
Let A = O 2 , and define T: !Jl2 ➔ !Jl2 be defined by .

T(x) =Ax, find the image under Tofu =[ -~J and v = [:] . [Sl

7. Let A = _3 [2 1SJ and B =[43 -SJ

k . What :alue(s) of k, if any, will make AB = BA?

8. ~mputedetA,wMreA=[f ~ ! ~]
 . Model Queatton
9. Let H be th [2t] •
. . e set of all vectors of.the fotm ~t . Show that His a subspace of !IV.
[51
10. Find basis and the dimension of the supspace

H-{["•;~1 I
s,te ~}.
r ' .
[51

11. Find the eigenvalues and eigenvectors of A=[; _36] •.. [5]

12. Define orthogonal set. Show that (u,, u,, u,} is an orthogonal set, where Ut = [n, 1], u, = [

tlJ =[ ~1:J. .., . 151

I
13. Let W = span (x1, x,}, where x, = [i] and x, = G] . Construct an orthogonal basis (v1, v,} for

~ r rlr i. ~,
~

14: Let* be defined on cz+ by a* b = a; . Then show that (2+ forms a group. [5]

15. Define ring with ~ exampl~. Comp~te the product in the given ring (12) (16) in Z1s. [5]
: _i 1 • I. I I I,

□□□
f

•I

• I I r
 MATHEMATICS-0

Tribhuvan University
Institute of Science and Technology
2075
Bachelor Level/ First Year/Second Semester/ Science . Full Mark: 80
Computer Science and Technology (M1H 163) P~s Marks: 32
Mathematics II Time: 3 hrs.
Candidates are required to give their answers in their own words as far as practicable.
. . .

Group A (10 x 3 = 30)

Attempt any THREE q.-estions. .., ,· • , ~
1. · When a "System of linear equation is consistent and inconsistent? .Give an example for each.
Test the consistency and solve: x + y + z = 4, x + 2y + 2z = 2, 2x + 2y + z = 5. (2 + 1 + 7)
2. What is the condition of a matrix to have an inverse? Find the inverse of the ·matrix

.·A =[
1-2 -lJ
~1 .: : in exists. · , . -. , ,
r· ,
i
•
I
I
I -- ...
,- i

r•
(2 + 8)
I •
\ .l

3. Define linearly independent set of vectors with an example. Show that. the vectors (1, -4, 3),
(0, 3, 1) and (3, -5, 4) are linearly independe~t. DC? thet form a b~sis? Justify . .· • , (2 + 5 + 3)
1 3· 5

4. Find the least-square solution of Ax = b for A =

, • rlf ·• , , ('
1 1 0
l , r
'
and b
- ..= [3]
.., '

5
r.. r. I r

7 (10)
. · ·•· 1 1 2
3
1 3 3
Group B
Attempt any t en questions: (10 >< 5 = 50)

5. Change into reduce echelon form of the matrix ( : -7 8

3 -6) (5)
-9 · 12
6. Define linear transformation with an example. Is a transformation T: R2 ➔ R3 defined by
T(x, y) = (3x + y, Sx + 7y, x + 3y) linear? Justify. (2 + 3)

7. Let A= (-1 -2)

5 9
(9 2)
and B = k _ . What value (s) of kif any will make AB= BA?
1
5

1 5 -6
8. Define determinant. Evaluate without expanding -1 -4 4
(1 + 4)
-2 -7 9

9.
oJ '
Define subspace of a vector space. Let H = {(:\ s t e R} Show that H ·
' lS
b f R3
a SU space O .

(1 + 4)
 10. Find the dimension of the null
space
•
_ and column space of A =z
(-3 6 -1 1 - 7
1 -2 2 .3 _1
J
5
_ (6 3 -8 2 -4 5 8-4
11. Find the eigen values of the matrbc O _2 0 )
5
1 0 -3
2 5
12. Find LU factorization of the matrix ( )
6 -7 5
13. Define group. Show that the set of all . t
14. Define ring with . an ·:e xam . C fm m ege~ Z forms group under addition operation.(1 + 4)
1
4 x Zu. p e. o pu~e the product in the given . ring '(-3, 5) (2, -4) in
15. State and prove the
v = (1, 1).
~ag~rean theoremtof
.. .I (2.5 + 2.5)
two ~ ectors and verify this for u = (1, -1) and
.. . ( ~ :'] I,, ➔ Ii( - I (3 + 2)

Tribhuvan University
'" ~ · ln$ti.tute of Science an:d:Technology··
; 2076
Bachelor Ifv~/ f irst Year/Second Semester/ Science Full Mark: 80
f '?~puter Science~-T~~ology ( ~ ;t6~) Pass Marks: 32
Mathematics II ' r I • ' '" ._ • , ~ • •
· - J • ~ · ,,,,...-, -: 1.,-. r r•· •~tul \ , Time:·3 hrs.
(NEW~9URSE)
Candidates are required to give their answers in tfreir;:.ow,n ,'l,liords as far as practicable.
The figures in the margin indicate full marks'_ ;· ~, ,;.1 • ' 1 I

I r 1 .
1
Group A (10 x'3 '4 '30) -. · t r •·

Attempt any THREE questions. . : -~ · ·, i - , ,1 , •• - 1 ,

1,. When a system of linear equation i~ co~istent. anp plCOnsistent? Give an example for each.
Test the consistency and solve the system of equations:
,
2y = 5, - x + y + Sz = 2, y + z = 0. x-
(:;; • · ,frtOlJ ~ ,r- , • ' .. [2 + 2 + 6]

2. What is the conditibn of a ~trix to have an inverse? find the inverse of the matrix.

A=
5
1
1- ·2]
3 ,·if it exists.
0
. . \ ,,

[2+ 8]
[ I I

i
4 -3 8

~- , Find the least-square solution of Ax = b for A = [

-61
-2
~ andb=
[:_!1] .
- 2-
I I

[10]

4. Let T is a linear transformation. Find the standard matrix of T such that

i. T: R2 ➔ R4 by T(ei) = (3, 1, 3, 1) and T(e2) = (- 5,2,0,0) where ei = (1, 0) and e2) = (0, 1).
31t · · 1 kwis
ii. T: R2 ➔ R4 rotates point as the origin through 2 radians counter c oc e.
 MATREMATICS-D

T: R ➔ R 4 is a vertical shear transformation that maps ei into ei

iii. 2
- 2ei but leaves vector
ei unchanged. . _ (4 + 3 +3]
Group B (10 x 5 = 50)
Attempt and TEN questions. ·

5. For what v~ue of h willy be in span {v1, v2, v3 ) = [ !~ l =1l Y] v2 = [ V2 = [ and Y

=v2=[ f}
6. Let us define a linear ~formation T: R2 ➔R 2 by T(x) =l ~ .t ][ :: ] = J r; 2

Findthe~g~~der.!o~~=[ :1 .}v=[ i }~du+;v =I:]. 1

.' .i c
7~ la A= [ -~ ~ ] and B = [ ! -: ]. Determine the value (s) of kif any will make AB= BA. [5]

8. Define determinant. Compute the

·
det:~~t with~~t expanding [ ~1i -4~ :2 ]. [1 + 4]

- [ 1 2 ,3 ]
9. · Define null space. Find the basis for the null space of the matrix A = ·. 2 · 3 i _ ,. [1 + 4]
"( • .. " • t f •

10. C ' Let B = {bi, bi} and C = {c1, c2} be bases for a 'vector space V, and suppose b'i = - C1 + 4c2 and
' bi =Sci - Xl- Find the change of coordinate matrix for a vector space and find [x]c for x:,; Shi + 3bi.
l _.i , [2.5 + 2.5]

11. Find the eigenvalues oi'~~ ~a~ [ J~

6 5
8 ~ 6 :] .: . ' ., [51

12. Find the QR factorization of the ~trj.~ [ i ,~\ ] [51

13. Define binary operation. Determine whether the binary operation * is. associative or

..c:OililDUta~ye or both ~here~ is d~fine~,.~ n Q by letting i, firy = ~,;v. [51

14. ·Show that the ring (Z. + 4,. 4) is an integral domain. _ [51

15. Find the v~r x d e t e ~ b;;the ~oordinate vect~r [xi:~ [·:s: Jwh~ ·
! , , _,

P= [ 1],[-} ],[-: 1
] [5]

I ::

• •I
 I

I
I
f
TU Bxemtuatlon
Tribhuvan University
Institute of Science and Technology
2078
Bachelor Level/ First Year/Second Semester/ Science Full Mark: BO
Computer Science and Technology (Mil-1 163) Pass Marks: 32
Mathematics II Time: 3 hrs.
Candidates are required to give their ~~ in their o~ words as far as practica_ble.
Group A
Attempt any THREE questions. 10•3-30
j'·· ~,1. Define system of linear equations. When a system of equations is consistent! Determine if
e system
-~.!' - 2z1 - 3x2 + 4"3 = 5 '
X2-2x1 =4
Xt + 3z2 - Z3 = 2
cy Define linear transformation with an example.

2
. LetA=[;l -t }v= [_Jb=[ ;}x~[~j
<I

- l

- 1 -
,. - I
\

and define a transformation T : R2 ➔ R2 by T(z) = Ax then . I .J '.:, J

(a) find T(v) ~- • 1 • a v1 . :- t .:..rii · u. • -1

I (b) Fm
J
I
. d z e R2 whose image under T is b. [! ,, ,i , _ r
3_ Find the L. u factorization off (' · I. .J f - ·;o ~ 1

2 4 -1 5 -2
-4 -5 3 -8 1
.. .J
.
I

I
(

c 2 -5 -4 1 8 1 .. · JI ••

,J - , . ,. r1

4.

10x5=50
Attempt any TEN questions. . A linearly independent, where
I
5. Determination the co umn f the matrtx are ° - ·

A=[~ ~ -1
5 8 vectors
0
4

l
• R7 are equal?• Given an example. Compute u + 3v, u - 12v,
+4
6. When two column m
where
 MATHEMATICS-II

7. Let 5

A=[ -01 ~]
and define T : R2 ➔ by T(x) = Ax, find th~ ·image under T of

u = [ _ \ ] and v = [ : ]
8. Find the eigen values of 5

[0~ 0~ -68]
2
l ..,.., .., ,1 I •
9. Define null space of a matrix A. Let

A =C: :: :J =- and v
5 ]
[ 3 ,
-2
The show that v is the null A.
10. Verify that lk, (-2)k, 3k are linearly independent signals. r 5
7 2] _. , .' "'r", . r
11. HA= [ .FindaformulaforAn,whereA=PDP-1 and r • 5
-4 1 ' ·

1 •.

12. Find a unit vector v of u = (1, - 2, 2, 3) I the direction of u. :, 5

13. Prove that the two vectors u and v are perpendiculcl! to each other if and only if the line
through u is perpendicular bisector ,of the line segment from - u,to v. 5
:--.... lrn,, ~ - ab
~i4~ Lefan operation* be defined on Q* by a* b = . Then show that Q* forms a group.
2 5

15. Define ring and show that set of real numbers with respect to addition and multiplication
• • • t ( ~

operation IS a nng. 2+3

r, - .
' I
□□□

- n