with Open Texts

LINEAR ALGEBRA
with Applications
Open Edition

PARTIAL STUDENT
SOLUTION MANUAL
VERSION 2019 – REVISION A

ADAPTABLE | ACCESSIBLE | AFFORDABLE

by W. Keith Nicholson
Creative Commons License (CC BY-NC-SA)
 a d v a n c i n g l e a r n i n g

Champions of Access to Knowledge

OPEN TEXT ONLINE

ASSESSMENT

All digital forms of access to our high-quality We have been developing superior online for-
open texts are entirely FREE! All content is mative assessment for more than 15 years. Our
reviewed for excellence and is wholly adapt- questions are continuously adapted with the
able; custom editions are produced by Lyryx content and reviewed for quality and sound
for those adopting Lyryx assessment. Access pedagogy. To enhance learning, students re-
to the original source files is also open to any- ceive immediate personalized feedback. Stu-
one! dent grade reports and performance statistics
are also provided.

SUPPORT INSTRUCTOR
SUPPLEMENTS

Access to our in-house support team is avail- Additional instructor resources are also freely
able 7 days/week to provide prompt resolution accessible. Product dependent, these supple-
to both student and instructor inquiries. In ad- ments include: full sets of adaptable slides and
dition, we work one-on-one with instructors to lecture notes, solutions manuals, and multiple
provide a comprehensive system, customized choice question banks with an exam building
for their course. This can include adapting the tool.
text, managing multiple sections, and more!

Contact Lyryx Today!

[email protected]
 a d v a n c i n g l e a r n i n g

Linear Algebra with Applications

Open Edition

BE A CHAMPION OF OER!
Contribute suggestions for improvements, new content, or errata:
A new topic
A new example
An interesting new question
A new or better proof to an existing theorem
Any other suggestions to improve the material
Contact Lyryx at [email protected] with your ideas.

CONTRIBUTIONS

Author
W. Keith Nicholson, University of Calgary

Lyryx Learning Team

Bruce Bauslaugh Martha Laflamme
Peter Chow Jennifer MacKenzie
Nathan Friess Tamsyn Murnaghan
Stephanie Keyowski Bogdan Sava
Claude Laflamme Ryan Yee

LICENSE

Creative Commons License (CC BY-NC-SA): This text, including the art and illustrations, are available
under the Creative Commons license (CC BY-NC-SA), allowing anyone to reuse, revise, remix and
redistribute the text.
To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/4.0/
 a d v a n c i n g l e a r n i n g

Linear Algebra with Applications

Open Edition

Base Text Revision History

Current Revision: Version 2019 — Revision A

• New Section on Singular Value Decomposition (8.6) is included.

• New Example 2.3.2 and Theorem 2.2.4. Please note that this will impact the numbering of subsequent
examples and theorems in the relevant sections.
2019 A
• Section 2.2 is renamed as Matrix-Vector Multiplication.
• Minor revisions made throughout, including fixing typos, adding exercises, expanding explanations,
and other small edits.

• Images have been converted to LaTeX throughout.

2018 B • Text has been converted to LaTeX with minor fixes throughout. Page numbers will differ from 2018A
revision. Full index has been implemented.

2018 A • Text has been released with a Creative Commons license.

Contents

1 Systems of Linear Equations 1

1.1 Solutions and Elementary Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Gaussian Elimination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Homogeneous Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 An Application to Network Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 An Application to Electrical Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.6 An Application to Chemical Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Supplementary Exercises: Chapter 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2 Matrix Algebra 13
2.1 Matrix Addition, Scalar Multiplication, and Transposition . . . . . . . . . . . . . . . . . . 13
2.2 Matrix-Vector Multiplication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Matrix Multiplication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4 Matrix Inverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.5 Elementary Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.6 Matrix Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.7 LU-factorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.8 An Application to Input-Output Economic Models . . . . . . . . . . . . . . . . . . . . . 36
2.9 An Application to Markov Chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Supplementary Exercises: Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3 Determinants and Diagonalization 41

3.1 The Cofactor Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2 Determinants and Matrix Inverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.3 Diagonalization and Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.4 An Application to Linear Recurrences . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.5 An Application to Systems of Differential Equations . . . . . . . . . . . . . . . . . . . . 55
3.6 Proof of the Cofactor Expansion Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Supplementary Exercises: Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

4 Vector Geometry 59
4.1 Vectors and Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

iii
iv CONTENTS

4.2 Projections and Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

4.3 More on the Cross Product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.4 Linear Operators on R3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.5 An Application to Computer Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Supplementary Exercises: Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

5 The Vector Space Rn 75

5.1 Subspaces and Spanning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2 Independence and Dimension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.3 Orthogonality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.4 Rank of a Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.5 Similarity and Diagonalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.6 Best Approximation and Least Squares . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.7 An Application to Correlation and Variance . . . . . . . . . . . . . . . . . . . . . . . . . 87
Supplementary Exercises: Chapter 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

6 Vector Spaces 89
6.1 Examples and Basic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.2 Subspaces and Spanning Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.3 Linear Independence and Dimension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.4 Finite Dimensional Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.5 An Application to Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.6 An Application to Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Supplementary Exercises: Chapter 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

7 Linear Transformations 103

7.1 Examples and Elementary Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.2 Kernel and Image of a Linear Transformation . . . . . . . . . . . . . . . . . . . . . . . . 106
7.3 Isomorphisms and Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
7.4 A Theorem about Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
7.5 More on Linear Recurrences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

8 Orthogonality 117
8.1 Orthogonal Complements and Projections . . . . . . . . . . . . . . . . . . . . . . . . . . 117
8.2 Orthogonal Diagonalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
8.3 Positive Definite Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
8.4 QR-Factorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
8.5 Computing Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
 CONTENTS v

8.6 Singular Value Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

8.7 Complex Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
8.8 An Application to Linear Codes over Finite Fields . . . . . . . . . . . . . . . . . . . . . . 129
8.9 An Application to Quadratic Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
8.10 An Application to Constrained Optimization . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.11 An Application to Statistical Principal Component Analysis . . . . . . . . . . . . . . . . . 135

9 Change of Basis 137

9.1 The Matrix of a Linear Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
9.2 Operators and Similarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
9.3 Invariant Subspaces and Direct Sums . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

10 Inner Product Spaces 151

10.1 Inner Products and Norms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
10.2 Orthogonal Sets of Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
10.3 Orthogonal Diagonalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
10.4 Isometries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
10.5 An Application to Fourier Approximation . . . . . . . . . . . . . . . . . . . . . . . . . . 163

11 Canonical Forms 165

11.1 Block Triangular Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
11.2 Jordan Canonical Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

A Complex Numbers 169

B Proofs 175

C Mathematical Induction 177

1. Systems of Linear Equations

1.1 Solutions and Elementary Operations

1. b. Substitute these values of x1 , x2 , x3 and x4 in the equation

2x1 + 5x2 + 9x3 + 3x4 = 2(2s + 12t + 13) + 5(s) + 9(−s − 3t − 3) + 3(t) = −1
x1 + 2x2 + 4x3 = (2s + 12t + 13) + 2(s) + 4(−s − 3t − 3) = 1
Hence this is a solution for every value of s and t.
2. b. The equation is 2x + 3y = 1. If x = s then y = 13 (1 − 2s) so this is one form of the general
solution. Also, if y = t then x = 12 (1 − 3t) gives another form.

4. Given the equation 4x − 2y + 0z = 1, take y = s and z = t and solve for x: x = 14 (2s + 3). This is the
general solution.
5. a. If a = 0, no solution if b 6= 0, infinitely many if b = 0.
b. If a 6= 0 unique solution x = b/a for all b.
h i
7. b. The augmented matrix is 10 21 01 .
 
1 1 0 1
d. The augmented matrix is 0 1 1 0 .
−1 0 1 2

8. b. A system with this augmented matrix is

2x − y = −1
−3x + 2y + z = 0
y + z = 3
h h i i h i h i
1 2 1 −3
9. b. 3 → 10 −22 −41 → 10 21 12 →
4 −1
1
0
0
1 2 .
Hence x = −3, y = 2.
h i h i h i h i h i
d. 34 45 −31 → 43 54 −31 → 31 41 −41 → 1
0
1
1
−4
13 → 1
0
0
1
−17
13 .
Hence x = −17, y = 13.
" #
1 2 1 0
     
2 1 1 −1 1 2 1 0 1 2 1 0
10. b. 1 2 1 0 → 2 1 1 −1 → 0 −3 −1 −1 → 0 1 1
3
1
3
3 0 −2 5 3 0 −2 5 0 −6 −5 5 0 0 −3 7
1
− 23 1
" # " #
1 0 3 1 0 0 9
−7
→ 0 1 1
3
1
3 → 0 1 0 10
9 . Hence x = 19 , y = 10
9, z= 3 .
0 0 1 − 73 0 0 1 − 73

1
2 Systems of Linear Equations

h i h i
3 −2 5 3 −2 5
11. b. −12 8 16 → 0 0 36 . The last equation is 0x + 0y = 36, which has no solution.

14. b. False. The system x + y = 0, x − y = 0 is consistent, but x = 0 = y is the only solution.

d. True. If the original system was consistent the final system would also be consistent because
each row operation produces a system with the same set of solutions (by Theorem 1.1.1).

16. The substitution gives

3(5x′ − 2y′ ) + 2(−7x′ + 3y′ ) = 5
7(5x′ − 2y′ ) + 5(−7x′ + 3y′ ) = 1
this simplifies to x′ = 5, y′ = 1. Hence x = 5x′ − 2y′ = 23 and y = −7x′ + 3y′ = −32.

17. As in the Hint, multiplying by (x2 + 2)(2x − 1) gives x2 − x + 3 = (ax + b)(2x − 1) + c(x2 + 2).
Equating coefficients of powers of x gives equations 2a + c = 1, −a + 2b = −1, −b + 2c = 3.
Solving this linear system we find a = − 19 , b = − 59 , c = 11
9.

19. If John gets $x per hour and Joe gets $y per hour, the two situations give 2x+3y = 24.6 and 3x+2y =
23.9. Solving gives x = $4.50 and y = $5.20.

1.2 Gaussian Elimination

1. b. No, No; no leading 1.

d. No, Yes; not in reduced form because of the 3 and the top two 1’s in the last column.
f. No, No; the (reduced) row-echelon form would have two rows of zeros.
" 0 −1 3 1 3 2 1 # " 0 1 −3 −1 −3 −2 −1 #
0 −2 6 1 −5 0 −1 −1 −11 −4 −3
2. b. 0 3 −9 2 4 1 −1 → 00 0
0
0
0 5 13 7 2
0 1 −3 −1 3 0 1
" 0 1 −3 0 8 2 2
# 0 "0
0
0
1 −3
0
0
6
8
2
2 2
2
#
0 0 0 1 11 4 3 0 0 0 1 11 4 3
→ 0 0 0 0 −42 −13 −13 → 0 0 0 0 0 1 1
0 0 0 0 6 2 2 0 0 0 0 3 1 1
0 1 −3 0 8 0 0 −3
" # " 0 1 0 0 0 0
#
0 0 0 1 11 0 −1 0 0 −1
→ 0 0 0 0 3 0 0 → 00 0
0
0
0
1
0 1 0 0
0 0 0 0 0 1 1 0 0 0 0 0 1 1

3. b. The matrix is already in reduced row-echelon form. The nonleading variables are parameters;
x2 = r, x4 = s and x6 = t.
The first equation is x1 − 2x2 + 2x4 + x6 = 1, whence x1 = 1 + 2r − 2s − t.
The second equation is x3 + 5x4 − 3x6 = −1, whence x3 = −1 − 5s + 3t.
The third equation is x5 + 6x6 = 1, whence x5 = 1 − 6t.
d. First carry the matrix to reduced row-echelon form.
" 1 −1 2 4 6 2 # " 1 0 4 5 5 1
# " 1 0 4 0 5 −4
#
0 1 2 1 −1 −1 0 1 2 1 −1 −1 0 1 2 0 −1 −2
0 0 0 1 0 1 → 0 0 0 1 0 1 → 0 0 0 1 0 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 1.2. Gaussian Elimination 3

The nonleading variables are parameters; x3 = s, x5 = t.

The first equation is x1 + 4x3 + 5x5 = −4, whence x1 = −4 − 4s − 5t.
The second equation is x2 + 2x3 − x5 = −2, whence x2 = −2 − 2s + t.
The third equation is x4 = 1.
i  
− 17
h i h i h i h
3 −1 0 1 2 −1 1 2 −1 1 2 −1 1 0
4. b. 2 −3 1 → 2 −3 1 → 0 −7 3 → 0 1 − 3 → .
7 0 1 − 37

Hence x = − 17 , y = − 37 .
d. hNote that the variables
i h in the second
i h equation are
i in the wrong order.
3 −1 2 3 −1 2 1 − 31 2

−6 2 −4 → 0 0 0 → 0 0 0 . 3

The nonleading variable y = t is a parameter; then x = 23 + 13 t = 31 (t + 2).

f. hAgain the order
i ofh the variables
i is reversed in the second equation.
2 −3 5 2 −3 5
−2 3 2 → 0 0 7 . There is no solution as the second equation is 0x + 0y = 7.
     
−2 3 3 −9 3 −4 1 5 1 −1 4 −4
5. b. 3 −4 1 5 → −2 3 3 −9 → −2 3 3 −9
−5 7 2 −14
  −5
 7 2 −14
 −5 7 2 −14
1 −1 4 −4 1 0 15 −21
→ 0 1 11 −17 → 0 1 11 −17 .
0 2 22 −34 0 0 0 0
Take z = t (the nonleading variable). The equations give x = −21 − 15t, y = −17 − 11t.
     
1 2 −1 2 1 2 −1 2 1 2 −1 2
d. 2 5 −3 1 → 0 1 −1 −3 → 0 1 −1 −3 .
1 4 −3 3 0 2 −2 1 0 0 0 7
There is no solution as the third equation is 0x + 0y + 0z = 7.
     
3 −2 1 −2 1 −1 3 5 1 −1 3 5
f. 1 −1 3 5 → 3 −2 1 −2 → 0 1 −8 −17
−1 1 1 −1 −1 1 1 −1 0 0 4 4
   
1 0 −5 −12 1 0 0 −7
→ 0 1 −8 −17 → 0 1 0 −9 . Hence x = −7, y = −9, z = 1.
0 0 1 1 0 0 1 1
       
1 2 −4 10 1 2 −4 10 1 2 −4 10 1 0 0 4
h. 2 −1 2 5 → 0 −5 10 −15 → 0 1 −2 3 → 0 1 −2 3 .
1 1 −2 7 0 −1 2 −3 0 0 0 0 0 0 0 0
Hence z = t, x = 4, y = 3 + 2t.

6. b. Label the rows of the augmented matrix as R1 , R2 and R3 , and begin the gaussian algorithm on
the augmented matrix keeping track of the row operations:
   
1 2 −3 −3 R1 1 2 −5 5 R2
1 3 −5 5 R2 → 0 1 −2 8 R2 − R1
1 −2 5 −35 R3 0 −4 8 −32 R3 − R1

At this point observe that R3 − R1 = −4(R2 − R1 ), that is R3 = 5R1 − 4R2 . This means that
equation 3 is 5 times equation 1 minus 4 times equation 2, as is readily verified. (The solution
is x1 = t − 11, x2 = 2t + 8 and x3 = t.)
" 1 −1 1 −1 0 # " 1 −1 1 −1 0 # " 1 −1 1 −1 0 #
−1 1 1 1 0 0 0 2 0 0 0 1 −1 1 0
7. b. 1 1 −1 1 0 → 0 2 −2 2 0 → 0 0 1 0 0
1 1 1 1 0 0 2 0 2 0 0 1 0 1 0
1 0 0 0 0 1 0 0 0 0
" # " #
0 1 −1 1 0 0 1 0 1 0
→ 0 0 1 0 0 → 0 0 1 0 0 . Hence x4 = t; x1 = 0, x2 = −t, x3 = 0.
0 0 1 0 0 0 0 0 0 0
4 Systems of Linear Equations

1 1 2 −1 4 1 1 2 −1 4 1 0 7 −7 8
" # " # " #
0 3 −1 4 2 0 3 −1 4 2 0 0 14 −14 14
d. 1 2 −3 5 0 → 0 1 −5 6 −4 → 0 1 −5 6 −4
1 1 −5 6 −3 0 0 −7 7 −7 0 0 −7 7 −7
1 0 7 −7 8 1 0 0 0 1 1 0 0 0 1
" # " # " #
0 1 −5 6 −4 0 1 −5 6 −4 0 1 0 1 1
→ 0 0 14 −14 14 → 0 0 1 −1 1 → 0 0 1 −1 1 .
0 0 −7 7 −7 0 0 0 0 0 0 0 0 0 0
Hence x4 = t; x1 = 1, x2 = 1 − t, x3 = 1 + t.
h i h i
8. b. a1 2b −15 → 10 2 −bab 5−1 +a .
 
−2−5b
h i
1 b −1 1 0
Case 1 If ab 6= 2, it continues → 0 1 5+a → 2−ab
5+a
.
2−ab 0 1 2−ab

The unique solution is x = −2−5b 5+a

2−abi , y = 2−ab .
h
Case 2 If ab = 2, it is 10 0b 5−1 + a . Hence there is no solution if a 6= −5. If a = −5, then
1 −2
h i
−1
b = −2
5 and the matrix is 0 0
5
0
. Then y = t, x = −1 + 25 t.
1 12 1 1
h i h b
i h b
i h b
i
a 1 1 1 1
d. 2 1 b → a 1 1 → 0 1 − a 1 − ab → 0 2 − a 2 − ab .
2 2 2 2 2
2 2
   
1 b b−1
1 2 1 0
Case 1 If a 6= 2 it continues: → 2
2−ab
→ 2−a
2−ab
.
0 1 2−a 0 1 2−a
b−1 2−ab
The unique solution: x = 2−a , y= 2−a .
1
h b
i
1
Case 2 If a = 2 the matrix is 2 2
. Hence there is no solution if b 6= 1.
h 0
1 1
0
i 2(1 − b)
1
If b = 1 the matrix is 0 2
0
2
0
, so y = t, x = 21 − 12 t = 12 (1 − t).
     
2 1 −1 a 1 0 −1 c 1 0 −1 c
9. b. 0 2 3 b → 0 2 3 b → 0 2 3 b →
1 0 −1 c 2 1 −1 a 0 1 1 a − 2c
     
1 0 −1 c 1 0 −1 c 1 0 0 b − 2a + 5c
0 1 1 a − 2c → 0 1 1 a − 2c → 0 1 0 3a − b − 6c .
0 2 3 b 0 0 1 b − 2a + 4c 0 0 1 b − 2a + 4c
Hence, for any values of a, b and c there is a unique solution x = −2a + b + 5c, y = 3a − b − 6c,
and z = −2a + b + 4c.
     
1 a 0 0 1 a 0 0 1 0 −ab 0
d. 0 1 b 0 → 0 1 b 0 → 0 1 b 0 .
c 0 1 0 0 −ac 1 0 0 0 1 + abc 0
   
1 0 −ab 0 1 0 0 0
Case 1 If abc 6= −1, it continues: → 0 1 b 0 → 0 1 0 0 .
0 0 1 0 0 0 1 0
Hence we have the unique solution x = 0, y = 0, z = 0.
 
1 0 −ab 0
Case 2 If abc = −1, the matrix is 0 1 b 0 , so z = t, x = abt, y = −bt.
0 0 0 0
Note: It is impossible that there is no solution here: x = y = z = 0 always works.
     
1 a −1 1 1 a −1 1 1 a −1 1
f. −1 a − 2 1 −1 → 0 2(a − 1) 0 0 → 0 a−1 0 0 .
2 2 a−2 1 0 2(a − 1) a −1 0 0 a −1
   
1 1 −1 1 1 1 0 0
Case 1 If a = 1 the matrix is 0 0 0 0 → 0 0 1 −1 ,
0 0 1 −1 0 0 0 0
so y = t, x = −t, z = −1.
Case 2 If a = 0 the last equation is 0x + 0y + 0z = −1, so there is no solution.
 1.2. Gaussian Elimination 5

Case3 If a 6= 1 and a 6=
 0, there is a uniquesolution:
1 − a1
  
1 a −1 1 1 a −1 1 1 0 0
0 a−1 0 0 → 0 1 0 0
1
→ 0 1 0 0 .
0 0 a −1 0 0 1 −a 0 0 1 − 1a
Hence x = 1 − a1 , y = 0, z = − 1a .
h i h i
2 1 −1 3 1 12 − 21 3
10. b. 0 0 0 0 → 0 0 0 0 ; rank is 1.
2

d. It is in row-echelon form; rank is 3.

   
0 0 1 0 0 1
f. 0 0 1 → 0 0 0 ; rank is 1.
0 0 1 0 0 0
       
−2 3 3 1 −1 4 1 −1 4 1 −1 4
11. b. 3 −4 1 → 3 −4 1 → 0 −1 −11 → 0 1 11 ; rank is 2.
−5 7 2 −5 7 2 0 2 22 0 0 0
       
3 −2 1 −2 1 −1 3 5 1 −1 3 5 1 −1 3 5
d. 1 −1 3 5 → 3 −2 1 −2 → 0 1 −8 −17 → 0 1 −8 −17 ;
−1 1 1 −1 −1 1 1 −1 0 0 4 4 0 0 1 1
rank is 3.
a2 a2 a2
     
1 1 2 1 1 2 1 1 2
f. 1 1−a 2 0 → 0 −a 0 −a2 → 0 a 0 a2 .
2 2−a 6−a 4 0 −a 2 − a 4 − 2a2 0 0 2−a 4 − a2
   
1 1 2 0 1 1 2 0
If a = 0 we get 0 0 0 0 → 0 0 1 2 ; rank = 2.
0 0 2 4
   0 0 0 0 
1 1 2 4 1 1 2 4
If a = 2 we get 0 2 0 4 → 0 1 0 2 ; rank = 2.
0 0 0 0 0 0 0 0
a2 a2
   
1 1 2 1 1 2
If a 6= 0, a 6= 2, we get 0 a 0 a2 → 0 1 0 a ; rank = 3.
0 0 2 − a 4 − a2 0 0 1 2+a
 
1 0 1
12. b. False. A = 0 1 1 .
0 0 0
 
1 0 1
d. False. A = 0 1 1 .
0 0 0

f. False. The system 2x−y = 0, −4x+2y = 0 is consistent, but the system 2x−y = 1, −4x+2y =
1 is not consistent.
h. True. A has 3 rows so there can be at most 3 leading 1’s. Hence the rank of A is at most 3.
   
1 a b+c 1 a b+c
14. b. We begin the row reduction 1 b c + a → 0 b − a a − b . Now one of b − a and c −
1 c a+b 0 c−a a−c

 is nonzero(by hypothesis) so that row provides the second leading 1 (its row becomes
a
0 1 −1 ). Hence further row operations give
   
1 a b+c 1 0 b+c+a
→ 0 1 −1 → 0 1 −1
0 0 0 0 0 0

which has the given form.

16. b. Substituting the coordinates of the three points in the equation gives
1+1+a+b+c = 0 a + b + c = −2
25 + 9 + 5a − 3b + c = 0 5a − 3b + c = −34
9 + 9 − 3a − 3b + c = 0 3a + 3b − c = 18
6 Systems of Linear Equations

     
1 1 1 −2 1 1 1 −2 1 1 1 −2
5 −3 1 −34 → 0 −8 −4 −24 → 0 1 1
2 3
3 3 −1 18 0 0 −4 24 0 0 1 −6
" 1
# " #
1 0 2 −5 1 0 0 −2
→ 0 1 1
2 3 → 0 1 0 6 .
0 0 1 −6 0 0 1 −6
Hence a = −2, b = 6, c = −6, so the equation is x2 + y2 − 2x + 6y − 6 = 0.

18. Let a, b and c denote the fractions of the student population in Clubs A, B and C respectively. The
4 2
new students in Club A arrived as follows: 10 of those in Club A stayed; 10 of those in Club B go to
2
A, and 10 of those in C go to A. Hence
4 2 2
a= 10 a + 10 b + 10 c

Similarly, looking at students in Club B and C.

1 7 2
b= 10 a + 10 b + 10 c

5 1 6
c= 10 a + 10 b + 10 c

Hence

−6a + 2b + 2c = 0
a − 3b + 2c = 0
5a + b − 4c = 0
" #
1 0 − 85 0
     
−6 2 2 0 1 −3 2 0 1 −3 2 0
1 −3 2 0 → 0 −16 14 0 → 0 1 − 87 0 → 0 1 − 87 0 .
5 1 −4 0 0 16 −14 0 0 0 0 0 0 0 0 0

Thus the solution is a = 58 t, b = 78 t, c = t. However a + b + c = 1 (because every student belongs to

exactly one club) which gives t = 52 . Hence a = 20 5 7
, b = 20 8
, c = 20 .

1.3 Homogeneous Equations

h i h i
1 0 1 0 1 0 1 1
1. b. False. A = 0 1 1 0 . d. False. A = 0 1 1 0 .
 
h i 1 0 0
1 0 0
f. False. A = 0 1 0 . h. False. A = 0 1 0 .
0 0 0

     
1 2 1 0 1 2 1 0 1 0 −9 0
2. b. 1 3 6 0 → 0 1 5 0 → 0 1 5 0 .
2 3 a 0 0 −1 a−2 0 0 0 a+3 0
Hence there is a nontrivial solution when a = −3: x = 9t, y = −5t, z = t.
     
a 1 1 0 1 1 −1 0 1 1 −1 0
d. 1 1 −1 0 → a 1 1 0 → 0 1 − a 1 + a 0 .
1 1 a 0 1 1 a 0 0 0 a+1 0
Hence if a 6= 1 and a 6= −1, there is a unique, trivial solution. The other cases are as follows:
 1.3. Homogeneous Equations 7

   
1 1 −1 0 1 1 0 0
a=1: 0 0 2 0 → 0 0 1 0 ; x = −t, y = t, z = 0.
0 0 2 0 0 0 0 0
   
1 1 −1 0 1 0 −1 0
a = −1 : 0 2 0 0 → 0 1 0 0 ; x = t, y = 0, z = t.
0 0 0 0 0 0 0 0

3. b. Not a linear combination. If ax + by + cz = v then comparing entries gives equations 2a + b +

c = 4, a + c = 3 and −a + b − 2c = −4. Now carry the coefficient matrix to reduced form:
   
2 1 1 4 1 0 1 0
1 0 1 3 → 0 1 −1 0
−1 1 −2 −4 0 0 0 1

Hence there is no solution.

d. Here, if aa + by + cz = v then comparing entries gives equations 2a + b + c = 3, a + c = 0 and
−a + b − 2c = 3. Carrying the coefficient matrix to reduced form gives
   
2 1 1 3 1 0 1 0
1 0 1 0 → 0 1 −1 3
−1 1 −2 3 0 0 0 0

so the general solution is a = −t, b = 3 + t and c = t. Taking t = −1 gives the linear combina-
tion v = a + 2y − z.

4. b. We must determine if x, y and z exist such that y = xa1 + ya2 + za3 . Equating entries here gives
equations −x + 3y + z = −1, 3x + y + z = 9, 2y + z = 2 and x + z = 6. Carrying the coefficient
matrix to reduced form gives
" −1 3 1 −1 # " 1 0 0 2 #
3 1 1 9 0 1 0 −1
0 2 1 2 → 0 0 1 4
1 0 1 6 0 0 0 0

so the unique solution is x = 2, y = −1 and z = 4. Hence y = 2a1 − a2 + 4a3 .

5. b. Carry the augmented matrix to reduced form:

   
1 2 −1 1 1 0 1 2 0 2 3 0
−1 −2 2 0 1 0 → 0 0 1 1 2 0
−1 −2 3 1 3 0 0 0 0 0 0 0

Hence the general solution is x1 = −2r − 2s − 3t, x2 = r, x3 = −s − 2t, x4 = s and x5 = t. In

 T
matrix form, the general solution x = x1 x2 x3 x4 x5 takes the form
 −2r − 2s − 3t   −2   −2   −3 
r 1 0 0
x= −s − 2t  = r 0 +s −1 +t  −2 
s 0 1 0
t 0 0 1

Hence x is a linear combination of the basic solutions.

d. Carry the augmented matrix to reduced form:
" 1 1 −2 −2 2 0 # " 1 0 0 1 0 0
#
2 2 −4 −4 1 0 0 1 −2 −3 0 0
1 −1 2 4 1 0 → 0 0 0 0 1 0
−2 −4 8 10 1 0 0 0 0 0 0 0
8 Systems of Linear Equations

 T
Hence the general solution x = x1 x2 x3 x4 x5 is
 −t   0   −1

2s + 3t 2 3
x= s  = s 1 +t  0 
t 0 1
0 0 0

Hence x is a linear combination of the basic solutions.

x+y = 1
6. b. The system 2x + 2y = 2 has nontrivial solutions with fewer variables than equations.
−x − y = −1
7. b. There are n − r = 6 − 1 = 5 parameters by Theorem 1.2.2.
d. The row-echelon form has four rows and, as it has a row of zeros, has at most 3 leading
1’s. Hence rank A = r = 1, 2 or 3 (r 6= 0, because A has nonzero entries). Thus there are
n − r = 6 − r = 5, 4 or 3 parameters.
9. b. Insisting that the graph of ax + by + cz + d = 0 (the plane) contains the three points leads to
three linear equations in the four variables a, b, c and d. There is a nontrivial solution by
Theorem 1.3.1.
11. Since the system is consistent there are n−r parameters by Theorem 1.2.2. The system has nontrivial
solutions if and only if there is at least one parameter, that is if and only if n > r.

1.4 An Application to Network Flows

1. b. There are five flow equations, one for each junction:

f1 − f2 = 25
f1 + f3 + f5 = 50
f2 + f4 + f7 = 60
− f3 + f4 + f6 = 75
f5 + f6 − f7 = 40
 1 −1 0 0 0 0 0 25   1 −1 0 0 0 0 0 25 
1 0 1 0 1 0 0 50 0 1 1 0 1 0 0 25
 0 1 0 1 0 0 1 60 → 0 1 0 1 0 0 1 60 
0 0 −1 1 0 1 0 75 0 0 −1 1 0 1 0 75
0 0 0 0 1 1 −1 40 0 0 0 0 1 1 −1 40
1 0 1 0 1 0 0 50 1 0 0 1 0 0 1 85
   
0 1 1 0 1 0 0 25 0 1 0 1 0 0 1 60
→ 0 0 −1 1 −1 0 1 35  →  0 0 1 −1 1 0 −1 −35 
0 0 −1 1 0 1 0 75 0 0 0 0 1 1 −1 40
0 0 0 0 1 1 −1 40 0 0 0 0 1 1 −1 40
1 0 0 1 0 0 1 85
 
0 1 0 1 0 0 1 60
→ 0 0 1 −1 0 −1 0 −75 
0 0 0 0 1 1 −1 40
0 0 0 0 0 0 0 0
If we use f4 , f6 , and f7 as parameters, the solution is
f1 = 85 − f4 − f7
 1.5. An Application to Electrical Networks 9

f2 = 60 − f4 − f7
f3 = −75 + f4 + f6
f5 = 40 − f6 + f7

2. b. The solution to (a) gives f1 = 55 − f4 , f2 = 20 − f4 + f5 , f3 = 15 − f5 . Closing canal BC means

f3 = 0, so f5 = 15. Hence f2 = 35 − f4 , so f2 ≤ 30 means f4 ≥ 5. Similarly f1 = 55 − f4 so
f1 ≤ 30 implies f4 ≥ 25. Hence the range on f4 is 25 ≤ f4 ≤ 30.
3. b. The road CD.

1.5 An Application to Electrical Networks

2. The junction and circuit rules give:

Left junction I1 − I2 + I3 = 0
Right junction I1 − I2 + I3 = 0
Top circuit 5I1 + 10I2 = 5
Lower circuit 10I2 + 5I3 = 10
       
1 −1 1 0 1 −1 1 0 1 −1 1 0 1 −1 1 0
5 10 0 5 → 0 15 −5 5 → 0 3 −1 1 → 0 1 −2 −1
0 10 5 10 0 10 5 10 0 2 1 2 0 2 1 2
    " 1 0 0 −1 #
1 0 −1 −1 1 0 −1 −1 5
→ 0 1 −2 −1 → 0 1 −2 −14 → 0 1 0 35 .
0 0 5 4 0 0 1 5 4
0 0 1 5

Hence I1 = − 15 , I2 = 3
5 and I3 = 4
5.

4. The equations are:

Lower left junction I1 − I5 − I6 = 0
Top junction I2 − I4 + I6 = 0
Middle junction I2 + I3 − I5 = 0
Lower right junction I1 − I3 − I4 = 0
Observe that the last of these follows from the others (so may be omitted).
Left circuit 10I5 − 10I6 = 10
Right circuit −10I3 + 10I4 = 10
Lower circuit 10I3 + 10I5 = 20
   
1 0 0 0 −1 −1 0 1 0 0 0 −1 −1 0
0 1 0 −1 0 1 0 0 1 0 −1 0 1 0
0 1 1 0 −1 0 0 0 0 1 1 −1 −1 0 
   
 →
 0 0 0 0 10 −1 10   0 0 0 0 1 −1 1 
0 0 −10 10 0 0 10 0 0 −1 1 0 0 1
0 0 10 0 10 0 20 0 0 1 0 1 0 2
   
1 0 0 0 −1 −1 0 1 0 0 0 −1 −1 0
0 1 0 −1 0 1 0 0 1 0 0 −2 0 −2
0 0 1 1 −1 −1 0 0 0 1 0 1 0 2 
   
→
 0 0 0 0 1 −1 1
→
  0 0 0 0 1 −1 1 
0 0 0 2 −1 −1 1 0 0 0 0 3 1 5
0 0 0 −1 2 1 2 0 0 0 1 −2 −1 −2
10 Systems of Linear Equations

   
1 0 0 0 −1 −1 0 1 0 0 0 0 −2 1
0 1 0 0 −2 0 −2 0 1 0 0 0 −2 0
0 0 1 0 1 0 2 0 0 1 0 0 1 1
   
→
 0 0 0 1 −2 −1 −2
→
  0 0 0 1 0 −3 0


0 0 0 0 1 −1 1 0 0 0 0 1 −1 1
0 0 0 0 3 1 5 0 0 0 0 0 4 2
 
1 0 0 0 0 0 2
 0 1 0 0 0 0 1 
1
 
0 0 1 0 0 0
→ 2
. Hence I1 = 2, I2 = 1, I3 = 12 , I4 = 23 , I5 = 32 , I6 = 21 .
 
3
 0 0 0 1 0 0 2 
3
0 0 0 0 1 0
 
2
1
0 0 0 0 0 1 2

1.6 An Application to Chemical Reactions

2. Suppose xNH3 + yCuO → zN2 + wCu + vH2 O where x, y, z, w and v are positive integers. Equating
the number of each type of atom on each side gives

N : x = 2z Cu : y = w
H : 3x = 2v O:y=v

Taking v = t these give y = t, w = t, x = 23 t and z = 12 x = 13 t. The smallest value of t such that there
are all integers is t = 3, so x = 2, y = 3, z = 1 and v = 3. Hence the balanced reaction is

2NH3 + 3CuO → N2 + 3Cu + 3H2 O

4. 15Pb(N3)2 + 44Cr(MnO4 )2 → 22Cr2O3 + 88MnO2 + 5Pb3 O4 + 90NO

Supplementary Exercises: Chapter 1

1. b. No. If the corresponding planes are parallel and distinct, there is no solution. Otherwise they
either coincide or have a whole common line of solutions.
 6 6 16

" 1 4 −1 1 2 # " 1 4 −1 1 2
# 1 0 10 10 10
 0 1 −4 1 1 
2. b. 31 −62 13 20 51 → 00 −10 −10
4 −1 −1
4 −1 −1 →  0 0 010 100 100 .
1 14 −5 2 3 0 10 −4 1 1
0 0 0 0 0
1
Hence x3 = s, x4 = t are parameters, and the equations give x1 = 10 (16 − 6s − 6t) and x2 =
1
10 (1 + 4s − t).
     
1 1 3 a 1 1 3 a 1 1 3 a
3. b. a 1 5 4 → 0 1 − a 5 − 3a 4 − a2 → 0 1 − a 5 − 3a 4 − a22 .
1 a 4 a 0 a−1 1 0 0 0 3(2 − a) 4 − a
   
1 1 3 1 1 1 3 0
If a = 1 the matrix is 0 0 2 3 → 0 0 1 1 , so there is no solution.
0 0 3 3 0 0 0 1
 1.6. An Application to Chemical Reactions 11

   
1 1 3 2 1 0 2 2
If a = 2 the matrix is 0 −1 −1 0 → 0 1 1 0 , so x = 2 − 2t, y = −t, z = t.
0 0 0 0 0 0 0 0
If a 6= 1 and a 6= 2 there is a unique solution.  
 " 1 1 3 2 −a+4 −5a+8
# " #

1 1 3 a a 1 0 a−1 a−1
1 0 0 3(a−1)
2 a2 −4
0 1−a 5 − 3a 4 − a2
2
→ 0 1 3a−5 a−1
a −4
a−1 → 0 1 3a−5
a−1 a−1 → 0 1 0 −a−2
3(a−1)
.
0 0 3(2 − a) 4−a a+2 a+2 a+2
0 0 1 3 0 0 1 3 0 0 0 3
8−5a −a−2 a+2
Hence x = 3(a−1)
, y= 3(a−1)
, z= 3 .

4. If R1 and R2 denote the two rows, then the following indicate how they can be interchanged using
row operations of the other two types:
         
R1 R1 + R2 R1 + R2 R2 R2
→ → → →
R2 R2 −R1 −R1 R1

Note that only one row operation of Type II was used — a multiplication by −1.

6. Substitute x = 3, y = −1 and z = 2 into the given equations. The result is

3 − a + 2c = 0 a − 2c = 3
3b − c − 6 = 1 that is 3b − c = 9
3a − 2 + 2b = 5 3a + 2b = 7

This
 system oflinear
 equations for a, b and
 c has uniquesolution:
   
1 0 −2 3 1 0 −2 3 1 0 −2 3 1 0 −2 3 1 0 0 1
0 3 −1 7 → 0 3 −1 7 → 0 1 −7 9 → 0 1 −7 9 → 0 1 0 2 .
3 2 0 7 0 2 6 −2 0 2 6 −2 0 0 20 −20 0 0 1 −1
Hence a = 1, b = 2, c = −1.
       
1 1 1 5 1 1 1 5 1 1 1 5 1 0 0 2
8. 2 −1 −1 1 → 0 −3 −3 −9 → 0 1 1 3 → 0 1 1 3 .
−3 2 2 0 0 5 5 15 0 0 0 0 0 0 0 0

Hence the solution is x = 2, y = 3 −t, z = t. Taking t = 3 − i gives x = 2, y = i, z = 3 − i, as required.

If the real system has a unique solution, the solution is real because all the calculations in the
gaussian algorithm yield real numbers (all entries in the augmented matrix are real).
2. Matrix Algebra

2.1 Matrix Addition, Scalar Multiplication, and

Transposition

1. b. Equating entries gives four linear equations: a − b = 2, b − c = 2, c − d = −6, d − a = 2. The

solution is a = −2 + t, b = −4 + t, c = −6 + t, d = t.
d. Equating coefficients gives: a = b, b = c, c = d, d = a. The solution is a = b = c = d = t, t
arbitrary.
h i h i h i h i h i h i h i h i
9 − 30 + 7 −14
2. b. 3 −13 − 5 62 + 7 −11 = −39 − 30 10 + 7
−7 = −3 − 10 − 7 = −20

d. [ 3 −1 2 ] − 2 [ 9 3 4 ] + [ 3 11 −6 ] = [ 3 −1 2 ]−[ 18 6 8 ]+[ 3 11 −6 ]
= [ 3 − 18 + 3 −1 − 6 + 11 2 − 8 − 6 ] = [ −12 4 −12 ]
2 T
   
0 −1 0 1 −2
f. 1 0 −4 = −1 0 4
−2 4 0 2 −4 0
h iT h i h i h i h i h i h i
2 1 1 −1 2 −1 1 −1 6 −3 2 −2 4 −1
h. 3 −1 0 −2 2 3 =3 1 0 −2 2 3 = 3 0 − 4 6 = −1 −6
h i h i
3 −1 15 −5
3. b. 5C − 5 2 0 = 10 0

d. B + D is not defined as B is 2 × 3 while D is 3 × 2.

h iT h iT h i
+3 1−1
f. (A +C)T = 20 + 2 −1 + 0 = 5 0
2 −1 = 5
0 −1
2

h. A − D is not defined as A is 2 × 2 while D is 3 × 2.

h i h i h i h i
4. b. Given 3A + 21 = 5A − 2 30 , subtract 3A from both sides to get 21 = 2A − 2 30 . Now
h i h i h i h i
add 2 0 to both sides: 2A = 1 + 2 30 = 81 . Finally, multiply both sides by 12 :
3 2
h i h i
4
A = 12 81 = 1 .
2

5. b. Given 2A − B = 5(A + 2B), add B to both sides to get

2A = 5(A + 2B) + B = 5A + 10B + B = 5A + 11B

Now subtract 5A from both sides: −3A = 11B. Multiply by − 13 to get A = − 11

3 B.

13
14 Matrix Algebra

4X + 3Y = A
6. b. Given , subtract the first from the second to get X +Y = B − A. Now subtract 3
5X + 4Y = B
times this equation from the first equation: X = A − 3(B − A) = 4A − 3B. Then X +Y = B − A
gives Y = (B − A) − X = (B − A) − (4A − 3B) = 4B − 5A.
Note that this also follows from the Gaussian Algorithm (with matrix constants):
h i h i h i
4 3 A 5 4 B 1 1 B−A
5 4 B → 4 3 A → 4 3 A
h i h i
1 1 B−A 1 0 4A − 3B
→ 0 −1 5A − 4B → 0 1 4B − 5A

 
7. b. Given 2X − 5Y = 1 2 let Y = T where T is an arbitrary 1 × 2 matrix. Then 2X = 5T +
1 2 so X = 25 T + 12 1 2 , Y = T . If T = s t , this gives X = 52 s + 12 25 t + 1 ,
       
 
Y = s t , where s and t are arbitrary.

8. b. 5[3(A − B + 2C) − 2(3C − B) − A] + 2[3(3A − B +C) + 2(B − 2A) − 2C]

= 5[3A − 3B + 6C − 6C + 2B − A] + 2[9A − 3B + 3C + 2B − 4A − 2C]
= 5[2A − B] + 2[5A − B +C]
= 10A − 5B + 10A − 2B + 2C
= 20A − 7B + 2C
h i
9. b. Write A = ac db . We want p, q, r and s such that
h i h i h i h i h i h i
a b 1 0 1 1 1 0 0 1 p+q+r q+s
c d =p 0 1 +q 0 0 +r 1 0 +s 1 0 = r+s p

Equating components give four linear equations in p, q, r and s:

p + q + r = a
q + s = b
r + s = c
p = d

The solution is p = d, q = 12 (a + b − c − d), r = 12 (a − b + c − d), s = 12 (−a + b + c + d).

11. b. A + A′ = 0
−A + (A + A′ ) = −A + 0 (add − A to both sides)
(−A + A) + A′ = −A + 0 (associative law
0 + A′ = −A + 0 (definition of − A)
A′ = −A (property of 0)
···
 a1 0 0
  b1 0 ··· 0 
0 a2 ··· 0 0 b2 ··· 0
13. b. If A =  .. .. ..  and B =  .. .. .. ,
. . . . . .
0 0 ··· an 0 0 ··· bn
a1 − b1 0 ··· 0
 
0 a2 − b2 ··· 0
then A − B =  .. .. ..  so A − B is also diagonal.
. . .
0 0 ··· an − bn
h i
s t
14. b. st 1 is symmetric if and only if t = st; that is t(s − 1) = 0; that is s = 1 or t = 0.
 2.2. Matrix-Vector Multiplication 15

d. This matrix is symmetric if and only if 2s = s, 3 = t, 3 = s + t; that is s = 0 and t = 3.

h i  h iT  h iT h i
15. b. 83 01 = 3AT + 2 10 02 = (3AT )T + 2 10 02 = 3A + 20 04 .
h i h i h i h i h i
Hence 3A = 83 01 − 20 04 = 63 −30 , so A = 13 63 −30 = 21 −10 .
h i  h iT h i
1 1 T T 1 0 1 −1
d. 4A − 9 −1 0 = (2A ) − 5 −1 2 = 2A − 5 0 2 .
h i h i h i
Hence 2A = −99 90 − 50 −5 10 = 4 14
−9 −10 .
h i h i
2 7
Finally A = 21 −94 14
10 = − 9 −5 .
2

16. b. We have AT = A as A is symmetric. Using Theorem 2.1.2: (kA)T = kAT = kA; so kA is

symmetric.

19. b. False. Take B = −A for any A 6= 0.

d. True. The entries on the main diagonal do not change when a matrix is transposed.
f. True. Assume that A and B are symmetric, that is AT = A and BT = B. Then Theorem 2.1.2
gives
(kA + mB)T = (kA)T + (mB)T = kAT + mBT = kA + mB
for any scalars k and m. This shows that the matrix kA + mB is symmetric.

20. c. If A = S +W as in (b), then AT = ST +W T = S −W . Hence A + AT = 2S and A − AT = 2W ,

so S = 12 (A + AT ) and W = 12 (A − AT ).

22. b. If A = [ai j ] then (kp)A = [(kp)ai j ] = [k(pai j )] = k[pai j ] = k(pA).

2.2 Matrix-Vector Multiplication

1. b. x1 − 3x2 − 3x3 + 3x4 = 5

8x2 + 2x4 = 1
x1 + 2x2 + 2x3 + = 2
x2 + 2x3 − 5x4 = 0
" 1 # " −2 # " −1 # " 1
# " 5
#
−1 0 1 −2 −3
2. b. x1 2 + x2 −2 + x3 7 + x4 0 = 8
3 −4 9 −2 12

3. b. By Definition 2.4:
h i x1
 h i h i h i h i
1 2 3 1 2 3 x1 + 2x2 + 3x3
Ax = 0 −4 5 x2 = x1 0 + x2 −4 + x3 5 = −4x2 + 5x3
x3
h i x1
 h i h i
1 2 3 1 · x1 + 2 · x2 + 3 · x3 x1 + 2x2 + 3x3
By Theorem 2.2.5: Ax = 0 −4 5
x2 = 0 · x1 + (−4) · x2 + 5 · x3 = −4x2 + 5x3
x3
16 Matrix Algebra

d. By Definition 2.4: " #

  x1          
3 −4 1 6 3 −4 1 6 3x1 − 4x2 + x3 + 6x4
x2
Ax = 0 2 1 5
x3 = x1 0 + x2 2 + x3 1 + x4 5 = 2x2 + x3 + 5x4
−8 7 −3 0 −8 7 −3 0 −8x1 + 7x2 − 3x3
x4
By Theorem 2.2.5:
 " x1
#    
3 −4 1 6 3 · x1 + (−4) · x2 + 1 · x3 + 6 · x4 3x1 − 4x2 + x3 + 6x4
x2
Ax = 0 2 1 5
x3 = 0 · x1 + 2 · x2 + 1 · x3 + 5 · x4 = 2x2 + x3 + 5x4
−8 7 −3 0 (−8) · x1 + 7 · x2 + (−3) · x3 + 0 · x4 −8x1 + 7x2 − 3x3
x4
     
1 −1 −4 −4 1 −1 −4 −4 1 0 −1 −2
5. b. 1 2 5 2 → 0 3 9→ 6 . Hence x = t − 2, y = 2 − 3t,
0 1 3 2
1 1 2
0   0 2
 6 4
   0 0 0 0
x −2 + t −2 1
z = t; that is y = 2 − 3t = + t −3 .
2
z t 0 1
   
−2 1
Observe that 2 is a solution to the given system of equations, and −3 is a solution to
0 1
the associated homogeneous system.
" 2 1 −1 −1 −1 # " 1 1 −2 1 −2
# " 1 0 1 0 1
# " 1 0 0 1 3
#
3 1 1 −2 −2 0 −1 3 1 3 0 1 −3 −1 −3 0 1 0 −4 −9
d. −1 −1 2 1 2 → 0 −2 7 1 4 → 0 0 1 −1 −2 → 0 0 1 −1 −2 .
−2 −1 0 2 3 0 1 −4 0 −1 0 0 −1 1 2 0 0 0 0 0

" x x#1 = 3"− 3t,−xt 2 =

Hence − 9, x#3 = t "− 2, x#4 = t,
# 4t " 3 −1
1
x2 −9 + 4t −9 4
so x3 = −2 + t = −2 +t 1 .
x4 t 0 1
3 −1
" # " #
−9 4
Here −2 is a solution to the given equations, and 1 is a solution to the associated
0 1
homogeneous equations.

6. To say that x0 and x1 are solutions to the homogeneous system Ax = 0 of linear equations means
simply that Ax0 = 0 and Ax1 = 0. If sx0 + tx1 is any linear combination of x0 and x1 , we compute:

A(sx0 + tx1 ) = A(sx0 ) + A(tx1 ) = s(Ax0 ) + t(Ax1 ) = s0 + t0 = 0

using Theorem 2.2.2. This shows that sx0 + tx1 is also a solution to Ax = 0.

8. b. The reduction of the augmented matrix is

" 1 −2 1 2  −3 + 2s − 5t 
3 −4
# " 1 −2 0 0 5 −3
#
s
3 6 −2 −3 −11 11 0 0 1 0 −2 −1
−2 4 −1 1 −8 7 → 0 0 0 1 0 0 so x =  −1 + 2t 
0
−1 2 0 3 −5 3 0 0 0 0 0 0
t
−3 2 −5
      
0 1 0
is the general solution. Hence x =  −1  + s  0 +t  2  is the desired expression.
0 0 0
0 0 1
h ih i h i h i
1 2 2 0 1 2
10. b. False. 2 4 −1 = 0 has a zero entry, but 2 4 has no zero row.
d. True. The linear combination
 x1 a1 + · · · + xn an equals Ax where, by Theorem 2.2.1,
A = a1 · · · an is the matrix with these vectors ai as its columns.
 2.2. Matrix-Vector Multiplication 17

 
h i h i 2
1 1 −1
f. False. If A = 2 and x =
2 then Ax = 14 , and this is not a linear combination
0
0
1
h i h i h i
of 12 and 12 because it is not a scalar multiple of 12 .
 
h i 1 h i
1 −1
h. False. If A = −1 1 −1 , there is a solution 2 for b = 00 . But there is no solution
1
1
h i h i x  h i
for b = 10 . Indeed, if −11 −11 −11 y = 10 then x − y + z = 1 and −x + y − z = 0.
z
This is impossible.
h i h i
11. b. If xy is reflected in the line y = x the result is xy ; see the diagram for Example 2.4.12. In
h i h i h ih i h i
other words, T xy = yx = 01 10 x
y . So T has matrix 0 1
1 0 .
h i h i
d. If xy is rotated clockwise through π2 the result is −xy ; see Example 2.2.14. Hence
h i h i h ih i h i
x y 0 1 x 0 1
T y = −x = −1 0 y so T has matrix −1 0 .
 
x
13. b. The reflection of y in the yz-plane keeps y and z the same and negates x. Hence
z

        
x −x −1 0 0 x −1 0 0
T y = y = 0 1 0 y , so the matrix is 0 1 0 .
z z 0 0 1 z 0 0 1

 
16. Write A = a1 a2 · · · an where ai is column i of A for each i. If b = x1 a1 + x2 a2 + · · · + xn an
 T
where the xi are scalars, then Ax = b by Theorem 2.2.1 where x = x1 x2 · · · xn ; that is x is
a solution to the system Ax = b.

18. b. We are given that x1 and x2 are solutions to Ax = 0; that is Ax1 = 0 and Ax2 = 0. If t is any
scalar then, by Theorem 2.2.2, A(tx1 ) = t(Ax1 ) = t0 = 0. That is, tx1 is a solution to Ax = 0.
   T
22. Let A = a1 a2 · · · an where ai is column i of A for each i, and write x = x1 x2 · · · xn
 T
and y = y1 y2 · · · yn . Then
 T
x+y = x1 + y1 x2 + y2 · · · xn + yn

Hence we have

A(x + y) = (x1 + y1 )a1 + (x2 + y2 )a2 + · · · + (xn + yn )an Definition 2.4

= (x1 a1 + y1 a1 ) + (x2 a2 + y2 a2 ) + · · · + (xn an + yn an ) Theorem 2.1.1
= (x1 a1 + x2 a2 + · · · + xn an ) + (y1 a1 + y2 a2 + · · · + yn an ) Theorem 2.1.1
= Ax + Ay Definition 2.4
18 Matrix Algebra

2.3 Matrix Multiplication

h i 2 3 1
 h i h i
1 −1 2 2−1−2 3−9+0 1−7+4 −1 −6 −2
1. b. 2 0 4
1 9 7 = 4+0−4 6+0+0 2+0+8 = 0 6 10
−1 0 2
 
3 0
d. [ 1 3 −3 ] −2 1 =[ 3−6+0 0 + 3 − 18 ]=[ −3 −15 ]
0 6
 
2
f. [ 1 −1 3 ]= [ 2 − 1 − 24 ] = [ −23 ]
1
−8
h ih i h i h i
2 −1 6−5 −3 + 3
h. 35 1
2 −5 3 = 10 − 10 −5 + 6 = 1 0
0 1
 ′   ′
aa′
   
a 0 0 a 0 0 aa + 0 + 0 0+0+0 0+0+0 0 0
j. 0 b 0 0 b ′ 0
′
= 0 + 0 + 0 0 + bb + 0 0 + 0 + 0′ =
′ 0 bb′ 0
0 0 c 0 0 c 0+0+0 0+0+0 0 + 0 + cc 0 0 cc′

2. b. A2 , AB, BC and C2 are all undefined. The other

 products are
h i h i −2 12 h i
BA = −11 42 −104 , B2 = −17 −66 , CB = 2 −6 , AC =
4
−2
10
−1 ,
1 6
 
2 4 8
CA = −1 −1 −5 .
1 4 2
h i h i
2a + a1 2b + b1 7 2
3. b. The given matrix equation becomes −a + 2a1 −b + 2b1 = −1 4 .
Equating coefficients gives linear equations

2a + a1 = 7 2b + b1 = 2
−a + 2a1 = −1 −b + 2b1 = 4

The solution is: a = 3, a1 = 1; b = 0, b1 = 2.

h i h i h i h i
2 8 2 2 2 6 0 0 0
4. b. A − A − 6I = 2 5 − 2 −1 − 0 6 = 0 0

h ih i h i h i 1 0

1 −1 −9 −16 −14 −17 −2 −1 −2
5. b. A(BC) = 0 1 5 1 = 5 1 = 3 1 0
2 1 = (AB)C
5 8
h i h i h i h i h i
a b 0 0 0 0 b 0 0 0
6. b. If A = c d then A 1 0 = 1 0 A becomes d 0 = a b whence b = 0 and
h i
a 0
a = d. Hence A has the form A = c a , as required.

7. b. If A is m × n and B is p × q then n = p because AB can be formed and q = m because BA can

be formed. So B is n × m, A is m × n.
h i h i h i h i h i h i
1 0 1 0 1 1 1 0 1 0 1 1
8. b. (i) 0 1 , 0 −1 , 0 −1 (ii) 0 0 , 0 1 , 0 0
h i h i h i h i
12. b. Write A = P0 XQ where P 10 −11 , X = 20 −10 , and Q = −10 1
1 . Then PX + X Q =
h i h i h i h i
2 −1
+ −2 1
= 0, so A 2 = P2 PX + XQ = P2 0 . Then
0 0 0 0 0 Q 2 0 Q 2
 2.3. Matrix Multiplication 19

h ih i h i h i
P2 0 P2 0 P4 0 P6 0
A4 = 0 Q2 0 Q2 = 0 Q4 , A6 = A4 A2 = 0 Q6 , . . . ; in general we claim that
h i
P2k 0
A2k = 0 Q2k for k = 1, 2, . . . (∗)

This holds for k = 1; if it holds for some k ≥ 1 then

h 2k ih 2 i h i
P2(k+1)
A2(k+1) = A2k A2 = P0 Q02k P
0
0
Q2 = 0
0
Q2(k+1)

Hence (∗) follows

h by
i induction
h in k.i
Next P = 0 1 , P = 10 −31 , and we claim that
2 1 −2 3

h i
1 −m
Pm = 0 1 for m = 1, 2, . . . (∗∗)

It is true for m = 1; if it holds for some m ≥ 1, then

h ih i h i
Pm+1 = Pm P = 10 −m1 1
0
−1
1 = 1
0
−(m + 1)
1

which proves (∗∗) by induction.

As to Q, Q2 = I so Q2k = I for all k. Hence (∗) and (∗∗) gives
" 1 −2k 0 0 #
h 2k i
0 1 0 0
A2k = P0 0I = 0 0 1 0 for k ≥ 1
0 0 0 1

Finally
1 −(2k + 1) 2 −1
" #
h ih i h i
P2k 0 P X P2k+1 P2k X 0 1 0 0
A2k+1 = A2k · A = 0 I 0 Q = 0 Q = 0 0 −1 1
0 0 0 1
h ih i h i h i
I X I −X I 2 + X0 −IX + XI I 0
13. b. 0 I 0 I = 0I + I0 −0X + I 2 = 0 I = I2k
h i
−X T
d. [ I XT ][ −X I ]T = [ I XT ] I = −IX T + X T I = Ok
h i2 h ih i h i
f. 0I X
0 = 0
I
X
0
0
I
X
0 = X
0
0
X
h i3 h ih i h i
0 X 0 X X 0 0 X2
I 0 = I 0 0 X = X 0
h i4 h ih i h 2 i
0 0 X2
I
X
0 = I
X
0
0
X 0 = X0 X02
h i2m h m i
Continue. We claim that 0I X
0 = X0 0
Xm for m ≥ 1. It is true if m = 1 and, if it holds
for some m, we have
h i2(m+1) h i2m h i2 h ih i h i
0 X Xm X m+1
I 0 = 0I X
0
0
I
X
0 = 0
0
Xm
X
0
0
X = 0
0
X m+1

Hence the result follows by induction on m. Now

h i2m+1 h i2m h i h ih i h i
0 X 0 X 0 X Xm 0 0 X 0 X m+1
I 0 = I 0 I 0 = 0 Xm I 0 = Xm 0

for all m ≥ 1. It also holds for m = 0 if we take X 0 = I.

20 Matrix Algebra

14. b. If YA = 0 for all 1 × m matrices Y , let Yi denote row i of Im . Then row i of Im A = A is Yi A = 0.

Thus each row of A is zero, so A = 0.

16. b. A(B + C − D) + B(C − A + D) − (A + B)C + (A − B)D = AB + AC − AD + BC − BA + BD −

AC − BC + AD − BD = AB − BA.
d. (A − B)(C − A) + (C − B)(A −C) + (C − A)2 = [(A − B) − (C − B) + (C − A)] (C − A) = 0(C −
A) = 0.

18. b. We are given that AC = CA, so (kA)C = k(AC) = k(CA) = C(kA), using Theorem 2.3.3. Hence
kA commutes with C.

20. Since A and B are symmetric, we have AT = A and BT = B. Then Theorem 2.3.3 gives (AB)T =
BT AT = BA. Hence (AB)T = AB if and only if BA = AB.
 
a x y
22. b. Let A = x b z . Then the entries on the main diagonal of A2 are a2 + x2 + y2 , x2 + b2 + z2 ,
y z c
y2 + z2 + c2 . These are all zero if and only if a = x = y = b = z = c = 0; that is if and only if
A = 0.

24. If AB = 0 where A 6= 0, suppose BC = I for some matrix C. Left multiply this equation by A to get
A = AI = A(BC) = (AB)C = 0C = 0, a contradiction. So no such matrix C exists.
" 1 0 1 0 # " 2 1 1 1 #
1 0 0 1 3 0 2 2
26. We have A = 0 0 0 1 , and hence A3 = 2 0 1 1 . Hence there are 3 paths of length 3
1 1 0 0 3 1 2 1
from v1 to v4 because the (4, 1)-entry of A3 is 3. Similarly, the fact that the (3, 2)-entry of A3 is 0
means that there are no paths of length 3 from v2 to v3 .
h i
27. b. False. If A = 10 00 = J then AJ = A, but J 6= I.
d. True. Since A is symmetric, we have AT = A. Hence Theorem 2.1.2 gives (I +A)T = I T +AT =
I + A. In other words, I + A is symmetric.
h i
f. False. If A = 00 10 then A 6= 0 but A2 = 0.
h. True. We are assuming that A commutes with A + B, that is A(A + B) = (A + B)A. Multiplying
out each side, this becomes A2 + AB = A2 + BA. Subtracting A2 from each side gives AB = BA;
that is A commutes
h with i B. h i
j. False. Let A = 21 42 and B = −21 −24 . Then AB = 0 is the zero matrix so both columns
are zero. However
h B has ino zero column.
h i
−2
l. False. Let A = −21 4 and B = 2
1
4
2 as above. Again AB = 0 has both rows zero, but A
has no row of zeros.

28. b. If A = [ai j ] the sum of the entries in row i is ∑nj=1 ai j = 1. Similarly for B = [bi j ]. If AB =
C = [ci j ] then ci j is the dot product of row i of A with column j of B, that is ci j = ∑nk=1 aik bk j .
Hence the sum of the entries in row i of C is
!
n n n n n n
∑ ci j = ∑ ∑ aik bk j = ∑ aik ∑ bk j = ∑ aik · 1 = 1
j=1 j=1 k=1 k=1 j=1 k=1
 2.3. Matrix Multiplication 21

Easier Proof: Let X be the n × 1 column matrix with every entry equal to 1. Then the entries
of AX are the row sums of A, so these all equal 1 if and only if AX = X . But if also BX = X
then (AB)X = A(BX ) = AX = X , as required.

30. b. If A = [ai j ] then the trace of A is the sum of the entries on the main diagonal, that is tr A =
a11 + a22 + · · · + ann . Now the matrix kA is obtained by multiplying every entry of A by k, that
is kA = [kai j ]. Hence

tr (kA) = ka11 + ka22 + · · · + kann = k(a11 + a22 + · · · + ann ) = k tr A

e. If A = [ai j ] the transpose AT is obtained by replacing each entry ai j by the entry a ji directly
across the main diagonal. Hence, write AT = [a′i j ] where a′i j = a ji for all i and j. Let bi denote
the (i, i)-entry of AAT . Then bi is the dot product of row i of A and column i of AT , that is
bi = ∑nk=1 aik a′ki = ∑nk=1 aik aik = ∑nk=1 a2ik . Hence we obtain
!
n n n n n
tr (AAT ) = ∑ bi = ∑ ∑ a2ik =∑ ∑ a2ik
i=1 i=1 k=1 i=1 k=1

This is what we wanted.

32. e. We have Q = P + AP − PAP so, since P2 = P,

PQ = P2 + PAP − P2 AP = P + PAP − PAP = P

Hence Q2 = (P + AP − PAP)Q = PQ + APQ − PAPQ = P + AP − PAP = Q.

34. b. We always have

(A + B)(A − B) = A2 + BA − AB − B2
If AB = BA, this gives (A + B)(A − B) = A2 − B2 . Conversely, suppose that (A + B)(A − B) =
A2 − B2 . Then
A2 − B2 = A2 + BA − AB − B2
Hence 0 = BA − AB, whence AB = BA.
   
35. b. Denote B = b1 b2 · · · bn = b j where b j is column j of B. Then Definition 2.9
asserts that    
AB = Ab1 Ab2 · · · Abn = Ab j ,
that is column j of AB is Ab j for each j. Note that multiplying a matrix by a scalar a is the
same as multiplying each column by a. This, with Definition 2.9 and Theorem 2.2.2, gives

a(AB) = a[Ab j ] Definition 2.9

= [a(Ab j )] Scalar Multiplication
= [A(ab j )] Theorem 2.2.2
= A(aB) Definition 2.9

Similarly,

a(AB) = a[Ab j ] Definition 2.9

22 Matrix Algebra

= [a(Ab j )] Scalar Multiplication

= [(aA)b j )] Theorem 2.2.2
= (aA)B Definition 2.9

This proves that a(AB) = A(aB) = (aA)B, as required.

36. See the article in the mathematics journal Communications in Algebra, Volume 25, Number 7
(1997), pages 1767 to 1782.

2.4 Matrix Inverses

2. In each case we need row operations that carry A to I; these same operations carry I to A−1 . In short
A I → I A−1 . This is called the matrix inversion algorithm.


b. hWe begin by subtracting

i h row 2 fromi rowh 1. i
4 1 1 0 1 −1 1 −1 1 −1 1 −1
| → →
3 2 0 1 3
i 
2 0 1
0 5 −3 4
2 1
h h i
1 −1 1 −1 1 0 −5 1 2 −1
→ 0 1 −3 4 → 5
3 4
. Hence the inverse is 5 −3 4 .
5 5 0 1 − 5 5
     
1 −1 2 1 0 0 1 −1 2 1 0 0 1 0 1 3 0 1
d. −5 7 −11 0 1 0 → 0 2 −1 5 1 0 → 0 1 −1 2 0 1
−2
 3 −5 0 0 1  0 1 −1 2 0 1
  0 0 1 1 1 −2
1 0 0 2 −1 3 2 −1 3
→ 0 1 0 3 1 −1 . So A−1 = 3 1 −1 .
0 0 1 1 1 −2 1 1 −2
     
3 1 −1 1 0 0 1 0 −1 1 −1 0 1 0 −1 1 −1 0
f. 2 1 0 0 1 0 → 2 1 0 0 1 0 → 0 1 2 −2 3 0
1 5 −1 0 0 1 1 5 −1 0 0 1 0 5 0 −1 1 1
 " 1 4 1
#

1 0 −1 1 −1 0 1 0 0 10 − 10 10

1 4 −1

1
→ 0 1 2 −2 3 0 → 0 1 0 2
− 10 2
10
2
10 . Hence A−1 = 10
−2 2 2 .
0 0 −10 9 −14 1 9 14 1 −9 14 −1
0 0 1 − 10 10 − 10

h. 
We begin by subtracting
 row 2 from twice row1:  
3 1 −1 1 0 0 1 0 −2 2 −1 0 1 0 −2 2 −1 0
5 2 0 0 1 0 → 5 2 0 0 1 0 → 0 2 10 −10 6 0
1 1 −1 0 0 1 1 1 −1 0 0 1 0 1 1 −2 1 1
2
 " − 24
# " #

1 0 −2 2 −1 0 1 0 −2 2 −1 0 1 0 0 4 0
→ 0 1 5 −5 3 0 → 0 1 5 −5 3 0 → 0 1 0 − 45 2
4
5
4 .
0 0 −4 3 −2 1 0 0 1 − 43 2
− 14
4 0 0 1 − 43 2
4 − 14
 
2 0 −2
1
Hence A−1 = 4
−5 2 5 .
−3 2 −1
−1 4 5 2 1 0 0 0 1 −4 −5 −2 −1 0 0 0
" # " #
0 0 0 −1 0 1 0 0 0 0 0 1 0 −1 0 0
j. 1 −2 −2 0 0 0 1 0 → 0 2 3 2 1 0 1 0
0 −1 −1 0 0 0 0 1 0 1 1 0 0 0 0 −1
1 −4 −5 −2 −1 0 0 0 1 0 −1 −2 −1 0 0 −4
" # " #
0 1 1 0 0 0 0 −1 0 1 1 0 0 0 0 −1
→ 0 2 3 2 1 0 1 0 → 0 0 1 2 1 0 1 2
0 0 0 1 0 −1 0 0 0 0 0 1 0 −1 0 0
 2.4. Matrix Inverses 23

1 0 0 0 0 0 1 −2 1 0 0 0 0 0 1 −2
" # " #
0 1 0 −2 −1 0 −1 −3 0 1 0 0 −1 −2 −1 −3
→ 0 0 1 2 1 0 1 2 → 0 0 1 0 1 2 1 2 .
0 0 0 1 0 −1 0 0 0 0 0 1 0 −1 0 0
0 0 1 −2
" #
−1 −2 −1 −3
Hence A−1 = 1 2 1 2 .
0 −1 0 0
 1 2 0 0 0 1 0 0 0 0   1 0 0 0 0 1 −2 6 −30 210 
0 1 3 0 0 0 1 0 0 0 0 1 0 0 0 0 1 −3 15 −105
l.  0 0 1 5 0 0 0 1 0 0 → 0 0 1 0 0 0 0 1 −5 35 .
0 0 0 1 1 0 0 0 1 0 0 0 0 1 0 0 0 0 1 −7
0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1
1 −2 6 −30 210
 
0 1 −3 15 −105
Hence A−1 =  0 0 1 −5 35 .
0 0 0 1 −7
0 0 0 0 1
h i h i h i
3. b. The equations are Ax = b where A = 21 −3−4 , x = x
y , b = 0
1 . We have (by the algo-
 
4 −3
rithm or Example 2.4.4) A−1 = 51 . Left multiply Ax = b by A−1 to get
1 −2
h ih i h i
x = A−1 Ax = A−1 b = 15 41 −3 −2
0
1 = 1 −3
5 −2

Hence x = − 35 and y = − 52 .
     
1 4 2 x 1
d. Here A = 2 3 3 , x = y , b = −1 .
4 1 4 0
 z 
9 −14 6
−1 1
By the algorithm, A = 5 4 −4 1 .
−10 15 −5
   " 23 #
9 −14 6 1 5
1
5
4 −4 1 −1 = 8
5
−10 15 −5 0 −5

The equations have the form Ax = b, so left multiplying by A−1 gives

    
9 −14 6 1 23
−1 −1 1 1
x = A (Ax) = A b = 5 9 −4 1 −1 =5 8
−10 15 −5 0 −25

23
Hence x = 5, y = 85 , and z = − 25
5 = −5.
 
1 −1 2
4. b. We want B such that AB = P where P = 0 1 1 . Since A−1 exists left multiply this
1 0 0
equation by A−1 to get B = A−1 (AB) = A−1 P. [This B will satisfy our requirements because
AB = A(A−1 P) = IP = P]. Explicitly
    
1 −1 3 1 −1 2 4 −2 1
−1
B=A P= 2 0 5 0 1 1 = 7 −2 4
−1 1 0 1 0 0 −1 2 −1

h i−1 h i
1 −1 1 3 1
5. b. By Example 2.4.4, we have (2A)T = 2 3 = 5 −2 1 . Since (2A)T = 2AT , we get
h i h i
2AT = 15 −23 11 so AT = 101 3
−2
1
1 . Finally
h iT h i
T T 1 3 1 1 3 −2
A = (A ) = 10 −2 1 = 10 1 1
24 Matrix Algebra

h i
2 1
d. We have (I − 2AT )−1 = 1 1 so (because (U −1 )−1 = U for any invertible matrix U )
h i−1 h i
2 1 1 −1
(I − 2AT ) = 1 1 = −1 2

h i h i h i h i
1 −1 1 0 1 −1 0 1
Thus 2AT =I− −1 2 = 0 1 − −1 2 = 1 −1 .
h i
1 0 1
This gives AT = 2 1 −1 , so
h iT h i
1 0 1 1 0 1
A = (AT )T = 2 1 −1 = 2 1 −1

h i −1 h i
1 0 1 0
f. Given 2 1 A = 2 2 , take inverses to get
h i h i−1 h i
1 0 1 0 1 2 0
2 1 A= 2 2 = 2 −2 1

h i−1 h i
1 0 1 0
Now 2 1 = −2 1 , so left multiply by this to obtain
h i h i h i
1 0 1 2 0 1 2 0
A= −2 1 2 −2 1 = 2 −6 1

h i
1 1
h. Given (A−1 − 2I)T = −2 1 0 , take transposes to get
 h iT h iT h i
1 1 1 1 1 1
A−1 − 2I = −2 1 0 = −2 1 0 = −2 1 0

h i h i h i h i h i
1 1 2 0 1 1 0 −2 0 −1
Hence A−1 = 2I − 2 1 0 = 0 2 −2 1 0 = −2 2 =2 −1 1 . Finally
 h i−1 h i−1  h i h i
−1 −1
A = (A−1 )−1 = 2 0
−1 1 = 1
2
0
−1 1 = 1
2
1
−1
1
1
1
0 = −1
2
1
1
1
0

 −1  
0 1 −1 2 −1 3
1
6. b. Have A = (A−1 )−1 = 1 2 1 = 2
0 1 −1 by the algorithm.
1 0 1 −2 1 −1
h i h
h i ih h i h i i
x′
8. b. The equations are A x
y where A = 34 45 and B = −54
= 7
1 and x
y =B y′
4
−3 .
h i h i h i
x′
Thus B = A−1 (by Example 2.4.4) so the substitution gives 71 = A xy = AB y′ =
h ′ i h ′ i h i h i h ih i h i
x x ′ ′ x 7 −5 4 7 −31
I y′ = y′ . Thus x = 7, y = 1 so y = B 1 = 4 −3 1 = 25 .
h i h i h i
1 0 1 0 2 0
9. b. False. A = 0 1 and B = 0 −1 are both invertible, but A + B = 0 0 is not.

d. True. If A4 = 3I then A( 13 A3 ) = I = ( 13 A3 )A, so A−1 = 31 A3 .

h i h i
f. False. Take A = 10 00 and B = 10 00 . Then AB = B and B 6= 0, but A is not invertible by
h i
Theorem 2.4.5 since Ax = 0 where x = 01 .
 2.4. Matrix Inverses 25

h. True. Since A2 is invertible, let (A2 )B = I. Thus A(AB) = I, so AB is the inverse of A by

Theorem 2.4.5.
10. b. We are given C−1 = A, so C = (C−1)−1 = A−1 . Hence CT = (A−1 )T . This also has the form
CT = (AT )−1 by Theorem 2.4.4. Hence (CT )−1 = AT .
11. b. If a solution x to Ax = b exists, it can be found by left multiplication by C: CAx = Cb, Ix = Cb,
x = Cb. h i h i
3
(i) x = Cb = 0 here but x = 30 is not a solution. So no solution exists.
h i
2
(ii) x = Cb = −1 in this case and this is indeed a solution.
h ih i h i h i
0 −1 0 −1 −1 0 1 0
15. b. B2 = 1 0 1 0= = 0 −1 so B4 (B2 )2 = 0 1 = I.
h ih i h i
3 3 −1 3 2 −1 0 0 −1 0 1
Thus B · B = I = B B, so B = B = B B = 0 −1 1 0 = −1 0 .

16. We use the algorithm:

     
1 0 1 1 0 0 1 0 1 1 0 0 1 0 1 1 0 0
c 1 c 0 1 0 → 0 1 0 −c 1 0 → 0 1 0 −c 1 0
3 c 2 0 0 1 0 c −1 −3 0 1 0 0 −1 c2 − 3 −c 1
−1
1 0 0 c2 − 2 −c c2 − 2
    
1 1 0 1 −c 1
→ 0 1 0 −c 2 1 0 . Hence c 1 c = −c 1 0 for all values of c.
0 0 1 3−c c −1 3 c 2 3 − c2 c −1

18. b. Suppose column j of A consists of zeros. Then Ay = 0 where y is the column with 1 in the
position j and zeros elsewhere. If A−1 exists, left multiply by A−1 to get A−1 Ay = A−1 0, that
is Iy = 0; a contradiction. So A−1 does not exist.
d. If each column of A sums to 0, then xA = 0 where x is the row of 1s. If A−1 exists, right
multiply by A−1 to get xAA−1 = 0A−1 , that is xI = 0, x = 0, a contradiction. So A−1 does not
exist.
 
2 1 −1
19. b. (ii) Write A = 1 1 0 . Observe that row 1 minus row 2 minus row 3 is zero. If x =
1 0 −1
1 −1 −1 , this means xA = 0. If A−1 exists, right multiply by A−1 to get xAA−1 =
 

0A−1 , xI = 0, x = 0, a contradiction. So A−1 does not exist.

20. b. If A is invertible then each power Ak is also invertible by Theorem 2.4.4. In particular, Ak 6= 0.
21. b. If A and B both have inverses, so also does AB (by Theorem 2.4.4). But AB = 0 has no inverse.
h i h i h ih i
2 2 x ax a 0 x
22. If a > 1, the x-expansion T : R → R is given by T y = y = 0 1 y . We have
h i−1 h 1 i
a 0
0 1 = 0a 10 , and this is an x-compression because a1 < 1.

24. b. The condition can be written as A(A3 + 2A2 − I) = 4I, whence A[ 14 (A3 + 2A2 − I)] = I. By
Corollary 2.4.1 of Theorem 2.4.5, A is invertible and A−1 = 41 (A3 + 2A2 − I). Alternatively,
this follows directly by verifying that also [ 14 (A3 + 2A2 − I)]A = I.
25. b. If Bx = 0 then (AB)x = 0 so x = 0 because AB is invertible. Hence B is invertible by Theorem
2.4.5. But then A = (AB)B−1 is invertible by Theorem 2.4.4 because both AB and B−1 are
invertible.
26 Matrix Algebra

h 2 −1
i
b. As in Example 2.4.11, −B−1YA−1 = −(−1)−1 1 3

26. −5 3 =[ −13 8 ], so

" #−1  " #−1  " #

3 1 0 3 1 0 2 −1 0
5 2 0 = h
5 2
i
0 = −5 3 0
1 3 −1 −13 8 (−1)−1 −13 8 −1

h ih ih i h i
1 −1 5 2 −2 1 −14 8
d. As in Example 2.4.11, −A−1 X B−1 = − −1 2 −1 0 −1 1 = 16 −9 , so

 −1  " #−1 " #   

2 1 5 2 2 1 −14 8 1 −1 −14 8
1 1 −1 0 1 1 16 −9 −1 2 16 −9
= =
   " # " #−1
  
0 0 1 −1 0 0 1 −1 0 0 2 −1
0 0 1 −2 0 0 1 −2 0 0 1 −1

28. d. If An = 0 write B = I + A + A2 + · · · + An−1 . Then

(I − A)B = (I − A)(I + A + A2 + · · · + An−1 )

= (I + A + A2 + · · · + An−1 ) − A − A2 − A3 − · · · − An
= I − An
=I

Similarly B(I − A) = I, so (I − A)−1 = B.

30. b. Assume that AB and BA are both invertible. Then

AB(AB)−1 = I so AX = I where X = B(AB)−1

(BA)−1BA = I so YA = I where Y = (BA)−1B

But then X = IX = (YA)X = Y (AX ) = Y I = Y , so X = Y is the inverse of A.

Different Proof. The fact that AB is invertible gives A[B(AB)−1] = I. This shows that A is
invertible by the Corollary 2.4.1 to Theorem 2.4.5. Similarly B is invertible.

31. b. If A = B then A−1 B = A−1 A = I. Conversely, if A−1 B = I left multiply by A to get AA−1 B = AI,
IB = A, B = A.

32. a. Since A commutes with C, we have AC = CA. Left-multiply by A−1 to get C = A−1CA. Then
right-multiply by A−1 to get CA−1 = A−1C. Thus A−1 commutes with C too.

33. b. The condition (AB)2 = A2 B2 means ABAB = AABB. Left multiplication by A−1 gives BAB =
ABB, and then right multiplication by B−1 yields BA = AB.

34. Assume that AB is invertible; we apply Part 2 of Theorem 2.4.5 to show that B is invertible. If
Bx = 0 then left multiplication by A gives ABx = 0. Now left multiplication by (AB)−1 yields
x = (AB)−1 0 = 0. Hence B is invertible by Theorem 2.4.5. But then we have A = (AB)B−1 so A is
invertible by Theorem 2.4.4 (B−1 and AB are both invertible).
 
−1
35. b. By the hint, Bx = 0 where x = 3 so B is not invertible by Theorem 2.4.5.
−1
 2.5. Elementary Matrices 27

36. Assume that A can be left cancelled. If Ax = 0 then Ax = A0 so x = 0 by left cancellation. Thus
A is invertible by Theorem 2.4.5. Conversely, if A is invertible, suppose that AB = AC. Then left
multiplication by A−1 yields A−1 AB = A−1 AC, IB = IC, B = C.

38. b. Write U = In − 2X X T . Then U is symmetric because

U T = InT − 2(X X T )T = In − 2X T T X T = In − 2X X T = U

Moreover U −1 = U because (since X T X = In )

U 2 = (In − 2X X T )(I − 2X X T )
= In − 2X X T − 2X X T + 4X X T X X T
= In − 4X X T + 4X Im X T
= In

39. b. If P2 = P then I − 2P is self-inverse because

(I − 2P)(I − 2P) = I − 2P − 2P + 4P2 = I

Conversely, if I − 2P is self-inverse then

I = (I − 2P)2 = I − 4P + 4P2

Hence 4P = 4P2 ; so P = P2 .

41. b. If A and B are any invertible matrices (of the same size), we compute:

A−1 (A + B)B−1 = A−1 AB−1 + A−1 BB−1 = B−1 + A−1 = A−1 + B−1

Hence A−1 + B−1 is invertible by Theorem 2.4.4 because each of A−1 , A + B, and B−1 is
invertible. Furthermore

(A−1 + B−1 )−1 = [A−1 (A + B)B−1 ]−1 = (B−1 )−1 (A + B)−1 (A−1 )−1 = B(A + B)−1A

gives the desired formula.

2.5 Elementary Matrices

1. b. Interchange rows 1 and 3 of I, E −1 = E.

 
1 0 0
d. Add (−2) times row 1 of I to row 2. E −1 = 2 1 0 .
0 0 1
 
1 0 0
−1
f. Multiply row 3 of I by 5. E = 0 1 0 .
1
0 0 5
28 Matrix Algebra

h i
−1 0
2. b. A → B is accomplished by negating row 1, so E = 0 0 .
h i
1 −1
d. A → B is accomplished by subtracting row 2 from row 1, so E = 0 1 .
h i
f. A → B is accomplished by interchanging rows 1 and 2, so E = 01 1
0 .
h i h i h i h i h i
0 1 k 0 1 0 1 k 1 0
3. b. The possibilities for E are 1 0 , 0 1 , 0 k , 0 1 and k 1 . In each case EA
has a row different from C.

4. If E is Type I, EA and A differ only in the interchanged rows.

If E is of Type II, EA and A differ only in the row multiplied by a nonzero constant.
If E is of Type II, EA and A differ only in the row to which a multiple of a row is added.

5. b. No. The zero matrix 0 is not invertible.

h i h i h i h 12
i
1 2 1 1 0 1 2 1 1 0 1 2 1 1 0 1 0 7 −1
6. b. 5 12 −1 0 1 → 0 2 −6 −5 1 → 0 1 −3 − 5 1 → 0 1 −3 − 5 1 so 2

h i h i 2 2 2 2
1 0 7 1 12 −2
UA = R = 0 1 −3 where U = 2 −5 1 . This matrix U is the product of the elementary
matrices used at each stage:
h i
1 2 1
5 12 −1 =A
↓
h i h i
1 2 1 1 0
0 2 −6 = E1 A where E1 = −5 1

↓
h i h i
1 2 1 1 0
0 1 −3 = E2 E1 A where E2 = 0 1
2

↓
h i h i
1 0 7 1 −2
0 1 −3 = E3 E2 E1 A where E3 = 0 1

d. Just as in (b), we get UA = R where R is reduced row-echelon, and

      
1 2 0 1 0 0 1 0 0 1 0 0 1 0 0 0 0 1
U= 0 1 0 1
0 5 0 0 1 0 0 1 0 −3 1 0 0 1 0
0 0 1 0 0 1 0 −1 1 −2 0 1 0 0 1 1 0 0

is a product of elementary matrices.

h i h i h i h i
E E
7. b. 21 −11 10 10 01 →1 12 −11 10 01 10 →2 32 −10 10 1
1
1
0 . So U = 1
1
1
0 .
h i h i
0 1 1 1
Then E1 = 1 0 and E2 = 0 1 , so U = E2 E1 .
 2.5. Elementary Matrices 29

8. b. h i
2 3
1 2 = A
h i ↓ h i
1 2 0 1
2 3 = E1 A where E1 = 1 0

h i ↓ h i
1 2 1 0
0 −1 = E2 E1 A where E2 = −2 1

h i ↓ h i
1 2 1 0
0 1 = E3 E2 E1 A where E3 = 0 −1

h i ↓ h i
1 0 1 −2
0 1 = E4 E3 E2 E1 A where E4 = 0 1

Thus E4 E3 E2 E1 A = I so

A = (E4 E3 E2 E1 )−1
= E1−1 E2−2 E3−1 E4−1
h ih ih ih i
= 01 10 1 0
2 1
1
0
0
−1
1
0
2
1

Of course a different sequence of row operations yields a different factorization of A.

    
1 0 0 1 0 0 1 0 −3 1 0 0
d. Analogous to (b), A = 0 1 0 0 1 0 0 1 0 0 1 4 .
−2 0 1 0 2 1 0 0 1 0 0 1

10. By Theorem 2.5.3, UA = R for some invertible matrix U . Hence A = U −1 R where U −1 is invertible.
 h i h i h i h i
12. b. A I = 32 21 10 01 → 12 11 10 −11 → 10 −11 −21 −13 → 10 01 −12 −32

h i
−1 2 −1
so U = 2 −3 . Hence, UA = R = I2 in this case so U = A . Thus, r = rank A = 2 and,
taking V = I2 , UAV = UA = I2 .
     
  1 1 0 −1 1 0 0 1 1 0 −1 1 0 0 1 0 1 3 −2 1 0
d. A I = 3 2 1 1 0 1 0 → 0 −1 1 4 −3 1 0 → 0 1 −1 −4 3 −1 0 .
0 −1 1 4 −1 0 1
1 0 1 3 0 0 1
   0 0 0 0 2 −1 1
−2 1 0 1 0 1 3
Hence, UA = R where U = 3 −1 0 and R = 0 1 −1 −4 . Note that rank A = 2.
2 −1 1 0 0 0 0
Next,
1 0 0 1 0 0 0 1 0 0 1 0 0 0 1 0 0 0
" # " # " #
 T 0 1 0 0 1 0 0 0 1 0 0 1 0 0 0 1 0 0
so V T =

R I = 1 −1 0 0 0 1 0 → 0 0 0 −1 1 1 0 −1 1 1 0 .
3 −4 0 0 0 0 1 0 0 0 −3 4 0 1 −3 4 0 1
1 0 0
" #  
1 0 0 0
0 1 0
Hence, (UAV )T = (RV )T = V T RT = 0 0 0 , so UAV = 0 1 0 0 .
0 0 0 0
0 0 0

16. We need a sequence of elementary operations to carry U A to I U −1 A . By Lemma 2.5.1

   

these operations can be achieved by left multiplication by elementary matrices. Observe

I U −1 A = U −1U U −1 A = U −1 U A
     
(2.1)
30 Matrix Algebra

Since U −1 is invertible, it is a product of elementary matrices (Theorem 2.5.2), sayU −1 = E1 E2 · · · Ek

where the Ei are elementary. Hence (2.1) shows that I U −1A = E1 E2 · · · Ek U A , so a se-
 

quence of k row operations carries U A to I U −1 A . Clearly I U −1A is in reduced

     

row-echelon form.
r r r
17. b. A ∼ A because A = IA. If A ∼ B, let A = U B, U invertible. Then B = U −1A so B ∼ A.
r r
Finally if A ∼ B and B ∼ C, let A = U B and B = VC where U and V are invertible. Hence
r
A = U (VC) = (UV )C so A ∼ C.
h i
19. b. The matrices row-equivalent to A = 00 00 01 are the matrices UA where U is invertible. If
h i h i
a b 0 0 b
U = c d then UA = 0 0 d where b and d are not both zero (as U is invertible). Every
h i
such matrix arises — use U = −ba ba — it is invertible as a2 + b2 6= 0 (Example 2.3.5).

22. b. By Lemma 2.5.1, B = EA where E is elementary, obtained from I by multiplying row i by

k 6= 0. Hence B−1 = A−1 E −1 where E −1 is elementary, obtained from I by multiplying row i
by 1k . But then forming the product A−1 E −1 is obtained by multiplying column i of A−1 by 1k .

2.6 Matrix Transformations

     
3 2 5
1. b. Write a = 2 ,b= 0 and x = 6 . We are given T (a) and T (b), and are asked to
−1 5 −13
find T (x). Since T is linear it is enough (by Theorem 2.6.1) to express x as a linear combination
of a and b. If we set x = ra + sb, equating entries gives equations 3r + 2s = 5, 2r = 6 and
−r + 5s = −13. The (unique) solution is r = 3, s = −2, so x = 3a − 2b. Since T is linear we
have h i h i h i
3 −1 11
T (x) = 3T (a) − 2T (b) = 3 5 − 2 2 = 11

1 −1 5
" # " # " #
1 1 −1
2. b. Let a = 1 ,b= 2 and x = 2 . We know T (a) and T (b); to find T (x) we express
1 −4 4
x as a linear combination of a and b, and use the assumption that T is linear. If we write
x = ra + sb, equate entries, and solve the linear equations, we find that r = 2 and s = −3.
Hence x = 2a − 3b so, since T is linear,
     
5 2 4
T (x) = 2T (a) − 3T (b) = 2 1 −3 0 = 2
−3 1 −9

h i h i
3. b. In R2 , we have e1 = 10 and e2 = 01 . We are given that T (x) = −x for each x in R2 . In
particular, T (e1 ) = −e and T (e2 ) = −e2 . Since T is linear, Theorem 2.6.2 gives
 h i
A = T (e1 ) T (e2 ) = −e1 −e2 = −10 −10
  
 2.6. Matrix Transformations 31

h i h i h i h ih i h i
x x −x −1 0 x x
Of course, T y =− y = −y = 0 −1 y in R2 , so in this case
for all y
h i
we can easily see directly that T has matrix −10 −10 . However, sometimes Theorem 2.6.2 is
necessary.
h i h i
d. Let e1 = 0 and e2 = 01 . If these vectors are rotated counterclockwise through π4 , some
1
 √   √ 
2
− 22
simple trigonometry shows that T (e1 ) = √2 and T (e2 ) =
2 √
2
. Since T is linear, the
2 2
h √ √ i
matrix A of T is A = T (e1 ) T (e2 ) = 12 √22 −√22 .
 

     
1 0 0
4. b. Let e1 = 0 , e2 = 1 and e3 = 0 denote the standard basis of R3 . Since T : R3 → R3
0 0 1
is reflection in the uz-plane, we have:
T (e1 ) = −e1 because e1 is perpendicular to the uz-plane; while
T (e2 ) = e2 and T (e3 ) = e3 because e2 and e3 are in the uz-plane.
 
    −1 0 0
So A = T (e1 ) T (e2 ) T (e3 ) = −e1 e2 e3 = 0 1 0 .
0 0 1

5. b. Since y1 and y2 are both in the image of T , we have y1 = T (x1 ) for some x1 in Rn , and
y2 = T (x2 ) for some x2 in Rn . Since T is linear, we have

T (ax1 + bx2 ) = aT (x1 ) + bT (x2 ) = ay1 + by2

This shows that ay1 + by2 = T (ax1 + bx2 ) is also in the image of T .

7. b. It turns out that T 2 fails for T : R2h→ R 2 2

i . T 2 requires that T (ax) = aT (x) for all x in R and
all scalars a. But if a = 2 and x = 01 then
 h i h i h i h i h i h i
0 0 0 0 0 0
T 2 1 =T 2 = 4 , while 2T 1 =2 1 = 2

Note that T 1 also fails for this transformation T , as you can verify.
h i h i h ih i h i
x x+y 1 1 x
8. √1
b. We are given T y = 2 −x + y = 2 −1 1 √1
y for all xy , so T is the matrix
h i  √1 √1 
1 1 1
transformation induced by the matrix A = √ −1 1 = − √12 √12 . By Theorem 2.6.4
2 2 2

we recognize this as the matrix of the rotation R− π . Hence T is rotation through θ = − π4 .

4
h i h i h ih i h i
x 8x + 6y −8 −6 x
1
d. Here T y = − 10 6x − 8y = 10 −6 8 1
y for all xy , so T is the matrix transfor-
h i
1 −8 −6
mation induced by the matrix A = 10 −6 8 . Looking at Theorem 2.6.5, we see that A is
the matrix of Q−3 . Hence T = Q−3 is reflection in the line y = −3x.
     
x 0 x
10. b. Since T is linear, we have T y = T y + T 0 . Since T is rotation about the y axis,
    z
  0 z
0 0 0
we have T y = y because y is on the y axis. Now observe that T is rotation of the
0 0 0
32 Matrix Algebra

xz-plane through the angle θ from the x axis to the z axis. By Theorem 2.6.4 the effect of T on
the xz-plane is given by
h i h ih i h i
x cos θ − sin θ x x cos θ − z sin θ
z → sin θ cos θ z = x sin θ + z cos θ
   
x x cos θ − z sin θ
Hence T 0 = 0 , and so
z x sin θ + z cos θ
         
x 0 x 0 x cos θ − z sin θ
T y =T y +T 0 = y + 0
z 0 z 0 x sin θ + z cos θ
    
x cos θ − z sin θ cos θ 0 − sin θ x
= y = 0 1 0 y
x sin θ + z cos θ sin θ 0 cos θ z
 
cos θ 0 − sin θ
Hence the matrix of T is 0 1 0 .
sin θ 0 cos θ

12. b. Let Q0 denoteh reflection

i h π denotei rotation through π . Then Q0 has
in the x axis, and let R
matrix A = 0 −1 , and Rπ has matrix B = −10 −10 . Then Rπ followed by Q0 is the
1 0
h i
−1 0
transformation Q0 ◦ Rπ , and this has matrix AB = 0 1 by Theorem 2.6.3. This is the
matrix of reflection in the y axis.
d. Let Q0 denote reflection in the x axis, and let R π denote rotation through π2 . Then Q0 has
h i h2 i
1 0 0 −1
matrix A = 0 −1 , and R π has matrix B = 1 0 . Then Q0 followed by R π is the
2 h i 2
transformation R π ◦ Q0 , and this has matrix BA = 01 10 by Theorem 2.6.3. This is the
2
matrix of reflection Q1 in the line with equation y = x.
f. Let Q0 denote reflection
h in ithe x axis, and let Q1 denote
h reflection
i in the line y = x. Then
1 0 0 1
Q0 has matrix A = 0 −1 , and Q1 has matrix B = 1 0 . Then Q0 followed by Q1 is
h i
0 −1
the transformation Q1 ◦ Q0 , and this has matrix BA = 1 0 by Theorem 2.6.3. This is the
matrix of rotation R π about the origin through the angle π2 .
2

13. b. Since R has matrix A, we have R(x) = Ax for all x in Rn . By the definition of T we have
T (x) = aR(x) = a(Ax) = (aA)x
for all x in Rn . This shows that the matrix of T is aA.
14. b. We use Axiom T 2: T (−x) = T [(−1)x] = (−1)T (x) = −T (x).
17. b. The matrix of T is B, so T (x) = Bx for all x in Rn . Let B2 = I. Then
T 2 (x) = T [T (x)] = B[Bx] = B2 x = Ix = x = 1R2 (x) for all x in Rn .
Hence T 2 = 1Rn since they have the same effect on every column x.
Conversely, if T 2 = 1Rn then
B2 x = B(Bx) = T (T (x)) = T 2 (x) = 1R2 (x) = x = Ix for all x in Rn .
This implies that B2 = I by Theorem 2.2.6.
 2.6. Matrix Transformations 33

h i h i h i h i
1 0 0 1 0 −1 0 −1
18. The matrices of Q0 , Q1 , Q−1 and R π are 0 −1 , 1 0 , −1 0 and 1 0 , respectively.
2
We use Theorem 2.6.3 repeatedly: If S has matrix A and T has matrix B then S ◦ T has matrix AB.
h ih i h i
0 −1
b. The matrix of Q1 ◦ Q0 is 01 10 1
0 −1
0
= 1 0 , which is the matrix of R π .
2
h ih i h i
1 0 0 −1 0 −1
d. The matrix of Q0 ◦ R π is 0 −1 1 0 = −1 0 which is the matrix of Q−1 .
2

19. b. We have Pm [Qm (x)] = Pm (x) for all x in R2 because Qm (x) lies on the line y = mx. This means
Pm ◦ Qm = Pm .
 T  T
20. To see that T is linear, write x = x1 x2 · · · xn and y = y1 y2 · · · yn . Then:
 T 
T (x + y) = T x1 + y1 x2 + y2 · · · xn + yn
= (x1 + y1 ) + (x2 + y2 ) + · · · + (xn + yn )
= (x1 + x2 + · · · + xn ) + (y1 + y2 + · · · + yn )
= T (x) + T (y)
 T
T (ax) = T ( ax1 ax2 · · · axn )
= ax1 + ax2 + · · · + axn
= a(x1 + x2 + · · · + xn )
= a T (x)
   
Hence T is linear, so its matrix is A = T (e1 ) T (e2 ) · · · T (en ) = 1 1 · · · 1 by Theorem
2.6.2.
Note that this can be seen directly because
 x1   x1

x2   x2
T .
..
 = x1 + · · · + xn = 1 1 ··· 1  .
..

xn xn

 
so we see immediately that T is the matrix transformation induced by 1 1 · · · 1 . Note that
this also shows that T is linear, and so avoids the tedious verification above.

22. b. Suppose that T : Rn → R is linear. Let e1 , e2 , . . . , en be the standard basis of Rn , and write
T (e j ) = w j for each j = 1, 2,. . . , n. Note that each w j is in
 R.As T is linear, Theorem
 2.6.2
asserts that T has matrix A = T (e1 ) T (e2 ) · · · T (en ) = w1 w2 · · · wn .
 x1 
x2
Hence, given x =  .  in Rn , we have
..
xn

x1
 
  x2
T (x) = Ax = w1 w2 · · · wn  .
..
 = w1 x1 + w2 x2 + · · · + wn xn = w · x = Tw (x)
xn
34 Matrix Algebra

T
for all x in Rn where w = w1 w2 · · · wn . This means that T = Tw . This can also be


seen without Theorem 2.6.2: We have x = x1 e1 + x2 e2 + · · · + xn en so, since T is linear,

T (x) = T (x1 e1 + x2 e2 + · · · + xn en )
= x1 T (e1 ) + x2 T (e2 ) + · · · + xn T (en )
= x1 w1 + x2 w2 + · · · + xn wn
= w·x
= Tw (x)

for all x in Rn . Thus T = Tw .

T S
24. b. Given linear transformations Rn → Rm → Rk we are to show that (S ◦ T )(ax) = a (S ◦ T )(x)
for all x in Rn and all scalars a. The proof is a straight forward computation:

(S ◦ T )(ax) = S[T (ax)] Definition of S ◦ T

= S[aT (x)] T is linear
= a[S[T (x)]] S is linear
= a[(S ◦ T )(x)] Definition of S ◦ T

2.7 LU-factorization
       
2 4 2 1 2 1 1 2 1 1 2 1
1. b. 1 −1 3 → 1 −1 3 → 0 −3 2 → 0 1 − 32 = U.
−1 7 −7 −1 7 −7 0 9 −6 0 0 0
 
2 0 0
Hence A = LU where U is above and L = 1 −3 0 .
−1 9 1
−1 −3 1 0 −1 1 3 −1 0 1 1 3 −1 0 1
" # " # " #
1 4 1 1 1 0 1 2 1 0 0 1 2 1 0
d. 1 2 −3 −1 1 → 0 −1 −2 −1 0 → 0 0 0 0 0 = U.
0 −2 −4 −2 0 0 −2 −4 −2 0 0 0 0 0 0
−1 0 0 0
" #
1 1 0 0
Hence A = LU where U is as above and L = 1 −1 1 0 .
0 −2 0 1
2 2 −2 4 2 1 1 −1 2 1 1 1 −1 2 1
" # " # " #
1 −1 0 2 1 0 −2 1 0 0 0 1 − 21 0 0
f. 3 1 −2 6 3 → 0 −2 1 0 0 → 0 0 0 0 0 = U.
1 3 −2 2 1 0 2 −1 0 0 0 0 0 0 0
2 0 0 0
" #
1 −2 0 0
Hence A = LU where U is above and L = 3 −2 1 0 .
1 2 0 1

2. b. The reduction to row-echelon form requires two row interchanges:

     
0 −1 2 0 −1 2 −1 2 1
0 0 4 → 0 0 4 → 0 −1 2 → ···
−1 2 1 −1 2 1 0 0 4
 2.7. LU-factorization 35

The elementary
 matrices corresponding
  (in order) to the interchanges
 are
1 0 0 0 1 0 0 0 1
P1 = 0 0 1 and P2 = 1 0 0 , so take P = P2P1 = 1 0 0 .
0 1 0 0 0 1 0 1 0
We apply the LU -algorithm to PA:
       
−1 2 1 1 −2 −1 1 −2 −1 1 −2 −1
PA = 0 −1 2 → 0 −1 2 → 0 1 −2 → 0 1 −2 =U
0 0 4 0 0 4 0 0 4 0 0 1
 
−1 0 0
Hence PA = LU where U is as above and L = 0 −1 0 .
0 0 4
d. "
The reduction to#row-echelon
" 1 2 −3form #requires
" 1 2two row interchanges:
−1 −2 −3
3 0 0 0
# " 1 2 −3 0
# " 1 2 −3 0
#
2 4 −6 5 0 0 0 5 0 0 0 5 0 1 −2 −3 0 1 −2 −3
1 1 −1 3 → 0 −1 2 3 → 0 1 −2 −3 → 0 0 0 5 → 0 0 −2 4
2 5 −10 1 0 1 −4 1 0 0 −2 4 0 0 −2 4 0 0 0 5
The elementary matrices corresponding (in #order) to the interchanges are #
1 0 0 0 1 0 0 0 1 0 0 0
" # " "
0 0 1 0 0 1 0 0 0 0 1 0
P1 = 0 1 0 0 and P2 = 0 0 0 1 so P = P2P1 = 0 0 0 1 .
0 0 0 1 0 0 1 0 0 1 0 0
We apply
" the LU -algorithm
# to" PA:
−1 −2 3 0 1 2 −3 0 1 2 −3 0
# " #
1 1 −1 3 0 −1 2 3 0 1 −2 −3
PA = 2 5 −10 1 → 0 1 −4 1 → 0 0 −2 4
2 4 −6 5 0 0 0 5 0 0 0 5
1 2 −3 0 1 2 −3 0
" # " #
0 1 −2 −3 0 1 −2 −3
→ 0 0 1 −2 → 0 0 1 −2 =U
0 0 0 5 0 0 0 1
−1 0 0 0
" #
1 −1 0 0
Hence PA = LU where U is as above and L̇ = 2 1 −2 0 .
2 0 0 5

    " x1
#  
2 0 0 1 1 0 −1 y1
x2
3. b. Write L = 1 3 0 ,U= 0 1 0 1 , x= x3 , y= y2 . The system Ly = b
−1 2 1 0 0 0 0 y3
x4
2y1 = −2
is y1 + 3y2 = −1 and we solve this by forward substitution: y1 = −1, y2 =
−y1 + 2y2 + y3 = 1
x1 + x2 − x4 = −1
1
3 (−1−y 1 ) = 0, y3 = 1+y1 −2y 2 = 0. The system U x = y is x2 + x4 = 0 and
0 = 0
we solve this by back substitution: x4 = t, x3 = 5, x2 = −x4 = −t, x1 = −1 + x4 − x2 = −1 + 2t.
" 2 # " 8 − 2t #
8 6−t
d. Analogous to (b). The solution is: y = −1 ,x= −1 − t , t arbitrary.
0 t

5. If the rows in question are R1 and R2 , they can be interchanged thus:

h i h i h i h i h i
R1 R1 + R2 R1 + R2 R1 R2
R2 → R2 → −R1 → −R1 → R1

6. b. Let A = LU = L1U1 be LU -factorizations of the invertible matrix A. Then U and U1 have

no row of zeros so (being row-echelon) are upper triangular with 1’s on the main diagonal.
Thus L−1
1 L = U1U
−1 is both lower triangular (L−1 L) and upper triangular (U U −1 ) and so is
1 1
diagonal. But it has 1’s on the diagonal (U1 and U do) so it is I. Hence L1 = L and U1 = U .
36 Matrix Algebra

7. We proceed
h h on niwhere A and B are n × n. It is clear if n = 1. In
iby induction h general, writei
A = X A1 and B = Y B1 where A1 and B1 are lower triangular. Then AB = Xb +abA1Y A10B1
a 0 b 0

by Theorem 2.2.5, and A1 B1 is upper triangular by induction. Hence AB is upper triangular.

9. b. Let A = LU = L1U1 be two such factorizations. Then UU1−1 = L−1 L1 ; write this matrix as
D = UU1−1 = L−1 L1 . Then D is lower triangular (apply Lemma 2.7.1 to D = L−1 L1 ), and D is
also upper triangular (consider UU1−1). Hence D is diagonal, and so D = I because L−1 and L1
are unit triangular. Since A = LU , this completes the proof.

2.8 An Application to Input-Output Economic Models

     
.5 0 −.5 1 0 −1 1 0 −1
1. b. I − E = −.1 .1 −.2 → 0 1 −3 →
. The equilibrium price structure p
0 1 −3
−.4 −.1 .7 0 −1 3 0 0 0
 T
is the solution to (I − E)p = 0; the general solution is p = t 3t t .
" .5 0 −.1 −.1 # " 1 2 −2 2 # " 1 2 −2 2
#
−.2 .3 0 −.1 5 0 −1 −1 0 −10 9 −11
d. −E = −.1 −.2 .2 −.2 → −2 3 0 −1 → 0 7 −4 3 .
−.2 −.1 −.1 .4 −2 −1 −1 4 0 3 −5 8
Now add 3 times row 4 to row 2 to get:  
14
" 1 2 −2 2 # " 1 0 −14 28 # 
1 0 −14 28
 1 0 0 − 23
17
0 −1 −6 13 0 1 6 −13 0 1 6 −13 0 1 0 − 23
→ → → 
0 7 −4 3 0 0 −46 94 47 47
0 0 1 − 23  0 0 1 − 23

0 3 −5 8 0 0 −23 47 0 0 0 0 0 0 0 0
The equilibrium price structure p is the solution to (I − E)p = 0. The solution is
 T
p = 14t 17t 47t 23t .
 
0 0 1
2. Here the input-output matrix is E = 1 0 0 so we get
0 1 0
     
1 0 −1 1 0 −1 1 0 −1
I −E = −1 1 0 → 0 1 −1 → 0 1 −1
0 −1 1 0 −1 1 0 0 0

Thus the solution to (I − E)p is p1 = p2 = p3 = t. Thus all three industries produce the same output.
h i h i h i
1−a −b 1 − a −b bt
4. I − E = −1 + a b → 0 0 so the possible equilibrium price structures are p = (1 − a)t ,
h i
t arbitrary. This is nonzero for some t unless b = 0 and a = 1, and in that case p = 11 is a solution.
h i
If the entries of A are positive then p = 1 −b a has positive entries.
h i h i
.4 .8 −1 5 8 8
7. b. One such example is E = .7 .2 , because (I − E) = − 4 7 6 .
h i h i
−a −b
8. If E = ac db then I − E = 1−c 1 − d . We have det (I − E) = (1 − a)(1 − d) − bc = 1 −
(a + d) +h (ad − bc) = −1
i 1 − tr E + det E. If det (I − E) 6= 0 then Example 2.3.5 gives (I − E) =
1 1−d b −1 ≥ 0 if
det (I−E) c 1 − a . The entries 1 − d, b, c, and 1 − a are all between 0 and 1 so (I − E)
det (I − E) > 0, that is if tr E < 1 + det E.
 2.9. An Application to Markov Chains 37

 
3
9. b. If p = 2 then p > Ep so Theorem 2.8.2 applies.
1
 
3
d. If p = 2 then p > Ep so Theorem 2.8.2 applies.
2

2.9 An Application to Markov Chains

1. b. Not regular. Every power of P has the (1, 2)- and (3, 2)-entries zero.
h 1 i h i h i
2. b. I − P = −21 −11 → 10 −20 so (I − P)s = 0 has solutions s = 2tt . The entries of s sum
2    
2 1
h i
1 1
to 1 if t = 3 , so s = 31 is the steady state vector. Given s0 = 0 , we get s1 = Ps0 = 21 ,
3 2
   
3 5
s2 = Ps1 41 , s3 = Ps2 = 83 . So it is in state 2 after three transitions with probability 38 .
4 8
     
.6 −.1 −.5 1 −2 1 1 0 −1
d. I − P = −.2 .4 −.2 → 0 11 −11 → 0 1 −1 so (I − P)s = 0 has solution s =
−.4 −.3 .7 0 −11 11 0 0 0
  " 1 #  
t 3 1
t . The entries sum to 1 if t = 1
3 so the steady state vector is s = 1
3 . Given s 0 = 0 ,
t 1 0
      3
.4 .38 .350
s1 = Ps0 = .2 , s2 = Ps1 = .28 , s3 = Ps2 = .312 . Hence it is in state 2 after three
.4 .34 .338
transitions with probability .312.
1 0 − 85
     
.9 −.3 −.3 1 −3 2
f. I − P = −.3 .9 −.6 → 0 24 −21 → 0 1 − 8 , so (I − P)s = 0 has solution s =
7

−.6 −.6 .9 0 −24 21 0 0 0

  " 5 #  
5t 20 1
7t . The entries sum to 1 if t = 1 so the steady state vector is s = 7
. Given s0 = 0 ,
8t
20 20
8 0
      20
.1 .28 .244
s1 = Ps0 = .3 , s2 = Ps1 = .42 , s3 = Ps1 = .306 . Hence it is in state 2 after three
.6 .30 .450
transitions with probability .306.
 
.7 .1 .1
4. b. The transition matrix is P = .1 .8 .3 where the columns (and rows) represent the up-
.2 .1 .6
per, middle and lower classes respectively and, for example, the last column asserts that, for
children of lower class people, 10%become upperclass, 30% become  middle
 class and 60%
.3 −.1 −.1 1 −2 3 1 0 −1
remain lower class. Hence I − P = −.1 .2 −.3 → 0 5 −10 → 0 1 −2 . Thus
−.2 −.1 −.4 0 −5 10 0 0 0
  " 1
#
t 4
the general solution to (I − P)s = 0 is s = 2t , so s = 1
2 is the steady state solution.
t 1
4
Eventually, upper, middle and lower classes will comprise 25%, 50% and 25% of this society
respectively.
38 Matrix Algebra

 
1 1
6. Let States 1 and 2 be “late” and “on time” respectively. Then the transition matrix is P = 3
2
2
1 .
3 2

Here column 1 describes what happens if he was late one day: the two entries sum to 1 and the top
entry is twice the bottom entry by the information we are given.
 Column
 2 is determined similarly.
3
h 3 i
Now if Monday is the initial state, we are given that s0 = 1 . Hence s1 = Ps0 = 85 and
4

4 8
 
7
7 9
s2 = Ps1 = 169 . Hence the probabilities that he is late and on time Wednesdays are 16 and 16
16

respectively.

8. Let the states be the five compartments. Since each tunnel entry is equally likely,
 1 1 1

0 2 5 0 2
1 1
 3 0 0 4 0 
P= 1
3 0 2
5
1
4
1
2

 1 1 2 
0 2 5 4 0
1 1
3 0 5 0 0

a. Since he starts in compartment 1,

   2
  7 
 1
 0 5 75
1 23
0  3   0   120 
s0 =  0  , s1 = Ps0 =  1
3
 , s2 = Ps1 =  3
10
 , s3 = Ps2 =  69
200

0 
0
  7   53

30 300
0 1 1 29
3 15 150

7
Hence the probability that he is in compartment 1 after three moves is 75 .
b. The steady state vector s satisfies (I − P)s = 0. As
 1 1 1
  
1 −2 −5 0 −2 1 0 0 0 − 32
 − 13 1 0 − 41 0   0 1 0 0 −1 
(I − P) =  − 13 0 3
5 − 41 − 12 → 0 0 1 0 − 52 
− 21 − 15 1
   
0 2 0 0 0 0 1 −2
− 13 0 − 15 0 1 0 0 0 0 0

 3 
2
1 
so the steady state is s = 16
5 . Hence, in the long run, he spends most of his time in
4
2
5
compartment 3 (in fact 16 of his time).
h i h i h i h i h i
1− p q 1 q 1 (1 − p)q + qp 1 q 1 q
12. a. p 1−q · p+q p = p+q pq + (1 − q)p = p+q p . Since the entries of p+q p add
to 1, it is the steady state vector.
b. If m = 1
h i h i h i
q + p − p2 − pq q − q + pq + q2
1
p+q
q
p
q
p + 1−p−q
p+q
p
−p
−q
q = 1
p+q p − p_p2 + pq p + q − pq − q2
h i
1 (p + q)(1 − p) (p + q)q
= p+q (p + q)p (p + q)(1 − q)
=P
 2.9. An Application to Markov Chains 39

h i h i
q q p −q
In general, write X = p p and Y = −p q . Then PX = X and PY = (1 − p − q)Y .
m
Hence if Pm = 1
p+q X + (1−p−q)
p+q Y for some m ≥ 1, then

1 (1−p−q)m
Pm+1 = PPm = p+q PX + p+q PY
m
= 1
p+q X + (1−p−q)
p+q (1 − p − q)Y
m+1
= 1
p+q X + (1−p−q)
p+q Y

Hence the formula holds for all m ≥ 1 by induction.

Now 0 < p < 1 and 0 < q < 1 imply 0 < p + q < 2, so that −1 < (p + q − 1) < 1. Multiplying
through by −1 gives 1 > (1 − p − q) > −1, so (1 − p − q)m converges to zero as m increases.

Supplementary Exercises: Chapter 2

2. b. We have 0 = p(U ) = U 3 −5U 2 +11U −4I so that U (U 2 −5U +11I) = 4I = (U 2 −5U +11I)U .
Hence U −1 = 14 (U 2 − 5U + 11I).

4. b. If xh = xm , then y = k(y − z) = y + m(y − z), whence (k − m)(y − z) = 0. But the matrix

y − z 6= 0 (because y 6= z) so k − m = 0 by Example 2.1.7.

6. d. Using (c), I pq AIrs = ∑ni=1 ∑nj=1 ai j I pq Ii j Irs . Now (b) shows that I pq Ii j Irs = 0 unless i = q and
j = r, when it equals I ps . Hence the double sum for I pq AIrs has only one nonzero term — the
one for which i = q, j = r. Hence I pq AIrs = aqr I ps .

7. b. If n = 1 it is clear. If n > 1, Exercise 6(d) gives

aqr I ps = I pq AIrs = I pq Irs A

because AIrs = Irs A. Hence aqr = 0 if q 6= r by Exercise 6(b). If r = q then aqq I ps = I ps A is the
same for each value of q. Hence a11 = a22 = · · · = ann , so A is a scalar matrix.
3. Determinants and Diagonalization

3.1 The Cofactor Expansion

If A is a square matrix, we write det A = |A| for convenience.

1. b. Take 3 out of row 1, then subtract 4 times row 1 from row 2:

6 9 2 3 2 3
8 12 = 3 8 12 = 3 0 0 = 0

a+1 a 1 1
d. Subtract row 2 from row 1: a a−1 = a a−1 = (a − 1) − a = −1
f. Subtract 2 times row 2 from row 1, then expand along row 2:
2 0 −3 0 −4 −13
−4 −13
1 2 5 = 1 2 5 =− 3 0 = −39
0 3 0 0 3 0
0 a 0
b d
h. Expand along row 1: b c d = −a 0 0 = −a(0) = 0
0 e 0

j. Expand along row 1:

0 a b
a c a 0
a 0 c = −a b 0 +b b c = −a(−bc) + b(ac) = 2abc
b c 0

l. Subtract multiples of row 1 from rows 2, 3 and 4, then expand along column 1:
1 0 3 1 1 0 3 1
2 0 −2
2 2 6 0 0 2 0 −2
−1 0 −3 1 = 0 0 0 2 = 0 0 2 =0
1 0 −4
4 1 12 0 0 1 0 −4

n. Subtract multiples of row 4 from rows 1 and 2, then expand along column 1:
4 −1 3 −1 0 −9 7 −5
−9 7 −5
3 1 0 2 0 −5 3 −1
0 1 2 2 = 0 1 2 2 =− −5 3 −1
1 2 2
1 2 −1 1 1 2 −1 1
Again, subtract multiples of row 3 from rows 1 and 2, then expand along column 1:
4 −1 3 −1
0 25 13
3 1 0 2 25 13 −1 −5
0 1 2 2 =− 0 13 9 = 13 9 =− 13 9 = −(−9 + 65) = −56
1 2 2
1 2 −1 1

p. Keep expanding along row 1:

0 0 0 a
0 0 b  
0 0 b p 0 c
0 c q k = −a 0 v q = −a b d s = −ab(−cd) = abcd
d s t
d s t u

−1 3 1 −1 3 1 −1 3 1 −1 3 1
5. b. 2 5 3 = 0 11 5 =− 0 1 2 =− 0 1 2 = −(17) = −17
1 −2 1 0 1 2 0 11 5 0 0 −17

41
42 Determinants and Diagonalization

2 3 1 1 1 1 2 5 1 1 2 5 1 1 2 5
0 2 −1 3 0 2 −1 3 0 2 −1 3 0 1 −3 −9
d. 0 5 1 1 =− 0 5 1 1 =− 0 5 1 1 = 0 5 1 1 =
1 1 2 5 2 3 1 1 0 1 −3 −9 0 2 −1 3
1 1 2 5 1 1 2 5
0 1 −3 −9 0 1 −3 −9
0 0 16 46 = 0 0 1 −17 = 106
0 0 5 21 0 0 0 106

a b c a b c
6. b. Subtract row 1 from row 2: a+b 2b c+b = b b b = 0 by Theorem 3.1.2(4).
2 2 2 2 2 2

7. b. Take −2 and 3 out of rows 1 and 2, then subtract row 3 from row 2, then take 2 out of row 2:
−2a −2b −2c a b c a b c a b c
2p + x 2q + y 2r + z = −6 2p + x 2q + y 2r + z = −6 2p 2q 2r = −12 p q r = 12
3x 3y 3z x y z x y z x y z

8. b. First add rows 2 and 3 to row 1:

2a + p 2b + q 2c + r 3a + 3p + 3x 3b + 3q + 3y 3c + 3r + 3z a+ p+x b+q+y c+r +z
2p + x 2q + y 2r + z = 2p + x 2q + y 2r + z =3 2p + x 2q + y 2r + z
2x + a 2y + b 2z + c 2x + a 2y + b 2z + c 2x + a 2y + b 2z + c

Now subtract row 1 from rows 2 and 3, and then add row 2 plus twice row 3 to row 1, to get
a+ p+x b+q+y c+r +z 3x 3y 3z
=3 p−a q−b r−c =3 p−a q−b r−c
x− p y−q z−r x− p y−q z−r

Next take 3 out of row 1, and then add row 3 to row 2, to get
x y z x y z
=9 p−a q−b r−c =9 −a −b −c
−p −q −r −p −q −r

Now use row interchanges and common row factors to get

−p −q −r −a −b −c a b c
= −9 −a −b −c =9 −p −q −r =9 p q r
x y z x y z x y z

h i
1 1
9. b. False. The matrix A = has zero determinant, but no two rows are equal.
2 2
h i h i
2 0 1 0
d. False. The reduced row-echelon form of A = 0 1 is R = 0 1 , but det A = 2 while
det R = 1.
h i
1 1
f. False. A = 0 1 , det A = 1 = det AT .
h i h i
1 1 1 0
h. False. If A = 0 1 and B = 1 1 then det A = det B = 1. In fact, it is a theorem that
det A = det AT holds for every square matrix A.

10. b. Partition the matrix as follows and use Theorem 3.1.5:

1 2 0 3 0
−1 3 1 4 0 2 1 1  
1 2 −1 2
0 0 2 1 1 = −1 0 2 = 5 −1 = −5(−7) = 35
−1 3 3 1
0 0 −1 0 2 3 0 1
0 0 3 0 1
 3.1. The Cofactor Expansion 43

11. b. Use Theorem 3.1.5 htwice: i

A 0 0
X B 0 = det XA B0 det C = ( det A det B) det C = 2(−1)3 = −6
Y Z C
A X 0 h i
A X
d. 0 B 0 = det 0 B det C = ( det A det B) det C = 2(−1)3 = −6
Y Z C

14. b. Follow the Hint, take out the common factor in row 1, subtract multiples of column 1 from
columns 2 and 3, and expand along row 1:
x−1 −3 1 x−2 x−2 x−2 1 1 1
det 2 −1 x−1 = 2 −1 x−1 = (x − 2) 2 −1 x−1
−3 x+2 −2 −3 x+2 −2 −3 x+2 −2
1 0 0
−3 x−3
= (x − 2) 2 −3 x−3 = (x − 2) x+5 1
−3 x+5 1

= (x − 2)(−x2 − 2x + 12) = −(x − 2)(x2 + 2x − 12)

2 −1
15. b. If we expand along column 2, the coefficient of z is − 1 3 = −(6 + 1) = −7. So c = −7.

16. b. Compute det A by adding multiples of row 1 to rows 2 and 3, and then expanding along column
1:
1 x x 1 x x
x2 − 2 x2 + x
det A = −x −2 x = 0 x2 − 2 x2 + x = x2 − x x2 − 3
−x −x −3 0 x2 − x x2 − 3

= (x2 − 2)(x2 − 3) − (x2 + x)(x2 − x) = (x4 − 5x2 + 6) − x2 (x2 − 1) = 6 − 4x2

√
3
Hence det A = 0 means x2 = 2 = 46 , so x = ± 6
2 .
d. Expand along column 1, and use Theorem 3.1.4:
x y 0 0
x y 0 y 0 0
0 0
det A = 0
x
0
y
x y =x 0 x y −y x y 0 = x · x3 − y · y3
0 0 x 0 x y
y 0 0 x

= x4 − y4 = (x2 − y2 )(x2 + y2 ) = (x − y)(x + y)(x2 + y2 )

Hence det A = 0 means x = y or x = −y (x2 + y2 = 0 only if x = y = 0).

 x1   y1 
x2 y2  
21. Let x =  .. , y =  .. , and A = c1 · · · x + y · · · cn where x + y is in column j. Ex-
. .
xn yn
panding det A along column j we obtain
n
T (x + y) = det A = ∑ (xi + yi )ci j (A)
i=1
n n
= ∑ xi ci j (A) + ∑ yi ci j (A)
i=1 i=1
= T (x) + T (y)

where the determinant at the second step is expanded along column 1. Similarly, T (ax) = aT (x) for
any scalar a.
44 Determinants and Diagonalization

24. Suppose A is n × n. B can be found from A by interchanging the following pairs of columns: 1 and
n, 2 and n − 1, . . . . There are two cases according as n is even or odd:
Case 1. n = 2k. Then we interchange columns 1 and n, 2 and n − 1, . . . , k and k + 1, k interchanges
in all. Thus det B = (−1)k det A in this case.
Case 2. n = 2k + 1. Now we interchange columns 1 and n, 2 and n − 1, . . . , k and k + 2, leaving
column k fixed. Again k interchanges are used so det B = (−1)k det A.
Thus in both cases: det B = (−1)k det A where A is n × n and n = 2k or n = 2k + 1.
Remark: Observe that, in each case, k and 12 n(n − 1) are both even or both odd, so (−1)k =
1 1
(−1) 2 n(n−1) . Hence, if A is n × n, we have det B = (−1) 2 n(n−1) det A.

3.2 Determinants and Matrix Inverses

 1 0 3 0 3 1 
−
−1 1 0 1 0 −1
   
1 −3 −3
−1 2 1 2 1 −1
 
1. b. The cofactor matrix is  − − = −1 1 1 .
−1 1 0 1 0 −1
−2 6 4
 
 
−1 2 1 2 1 −1
−
1 0 3 0 3 1
 
1 −1 −2
The adjugate is the transpose of the cofactor matrix: adj A = . −3 1 6
−3 1 4

d. In computing the cofactor matrix, we use the fact that det 31 M = 91 det M for any 2 × 2 matrix
 

M. Thus the cofactor matrix is

 1 −1 2 1 2 2 1 2 −1 
9 −9 9
2 −1 2 −1 2 2
     
−3 6 6 −1 2 2
2 2 −1 2 −1 2
  1 1
 − 19 1
9 − 91 = 6 −3 6 = 2 −1 2
 2 −1 2 −1 2 2  9 6 6 −3
3 2 2 −1
 
1 2 2 −1 2 −1 2
9 − 91 1
9
−1 2 2 2 2 −1
 
−1 2 2
1
The adjugate is the transpose of the cofactor matrix: adj A = 3
2 −1 2 . Note that the
2 2 −1
cofactor matrix is symmetric here. Note also that the adjugate actually equals the original
matrix in this case.
2. b. We compute the determinant by first adding column 3 to column 2:
0 c −c 0 0 −c
−1 1
−1 2 −1 = −1 1 −1 = (−c) c 0 = (−c)(−c) = c2
c −c c c 0 c
This is zero if and only if c = 0, so the matrix is invertible if and only if c 6= 0.
d. Begin by subtracting row 1 from row 3, and then subtracting column 1 from column 3:
4 c 3 4 c 3 4 c −1
c −1
c 2 c = c 2 c = c 2 0 =1 2 0 =2
5 c 4 1 0 1 1 0 0
This is nonzero for all values of c, so the matrix is invertible for all c.
 3.2. Determinants and Matrix Inverses 45

f. Begin by subtracting c times row 1 from row 2:

1 c −1 1 c −1
1 − c2 1+c (1 + c)(1 − c) 1+c
c 1 1 = 0 1 − c2 1+c = 1 c = 1 c
0 1 c 0 1 c
Now take the common factor (1 + c) out of row 1:
1 c −1
1−c 1
c 1 1 = (1 + c) 1 c = (1 + c)[c(1 − c) − 1] = −(1 + c)(c2 − c + 1) = −(c3 + 1)
0 1 c
This is zero if and only if c = −1 (the roots of c2 − c + 1 are not real). Hence the matrix is
invertible if and only if c 6= −1.

3. b. det (B2C−1 AB−1CT ) = det B2 det C−1 det A det B−1 det CT
= ( det B)2 det1 C det A det1 B det C
= det B det A
= −2
1 1
4. b. det (A−1 B−1 AB) = det A−1 det B−1 det A det B = det A det B det A det B = 1.
Note that the following proof is wrong:

det (A−1 B−1 AB) = det (A−1 AB−1 B) = det (I · I) = det I = 1

The reason is that A−1 B−1 AB may not equal A−1 AB−1 B because B−1 A need not equal AB−1 .

6. b. Since C is 3 × 3, the same is true for C−1 , so det (2C−1) = 23 · det C−1 = det8 C . Now we
compute det C by taking 2 and 3 out of columns 2 and 3, subtracting column 3 from column
2:
2p −a + u 3u p −a + u u p −a u
det C = 2q −b + v 3v =6 q −b + v v =6 q −b v
2r −c + w 3w r −c + w w r −c w

Now take −1 from column 2, interchange columns 1 and 2, and apply Theorem 3.2.3:
p a u a p u a b c
det C = −6 q b v =6 b q v =6 p q r = 6 · 3 = 18
r c w c r w u v w

8 8
Finally det 2C−1 = det C = 18 = 49 .

7. b. Begin by subtracting row 2 from row 3, and then expand along column 2:
2b 0 4d 2b 0 4d
2b 4d b 2d b d
1 2 −2 = 1 2 −2 =2 a 2c =4 a 2c =8 a c
a+1 2 2(c − 1) a 0 2c

Interchange rows and use Theorem 3.2.3, to get

a c a b
= −8 b d = −8 c d = −8(−2) = 16

9 4 3 9
−1 −1 −5 5 2 −1 −21 21
8. b. x = 3 4
= −11 = 11 , y= 3 4
= −11 = 11
2 −1 2 −1
46 Determinants and Diagonalization

d. The coefficient matrix has determinant:

4 −1 3 0 −1 0
14 5
6 2 −1 = 14 2 5 = −(−1) 15 11 = 79
3 3 2 15 3 11
Hence Cramer’s rule gives
1 −1 3 1 −1 3
1 1 1 2 −1 12
x= 79
0 2 −1 = 79
0 2 −1 = 79 2 5 = 79
−1 3 2 0 2 5
4 1 3 4 1 3
1 1 1 6 −1 37
y= 79
6 0 −1 = 79
6 0 −1 = − 79 7 5 = − 79
3 −1 2 7 0 5
4 −1 1 4 −1 1
1 1 1 6 2 2
z= 79
6 2 0 = 79
6 2 0 = 79 7 2 = − 79
3 3 −1 7 2 0

1 1
 T
b. A−1 =
 
9. det A adj A = det A Ci j where Ci j is the cofactor matrix. Hence the (2, 3)-entry of
1 2 −1 1 2 −1
1 1 −1
A−1 is det A C32 . Now C32 = − 3 1 = −4. Since det A = 3 1 1 = 0 −5 4 =
0 4 7 0 4 7
−5 4 −4 4
4 7 = −51, the (2, 3) entry of A−1 is −51 = 51 .

10. b. If A2 = I then det A2 = det I = 1, that is ( det A)2 = 1. Hence det A = 1 or det A = −1.
d. If PA = P, P invertible, then det PA = det P, that is det P det A = det P. Since det P 6= 0 (as
P is invertible), this gives det A = 1.
f. If A = −AT , A is n × n, then AT is also n × n so, using Theorem 3.1.3 and Theorem 3.2.3,

det A = det (−AT ) = det [(−1)AT ] = (−1)n det AT = (−1)n det A

If n is even this is det A = det A and so gives no information about det A. But if n is odd it
reads det A = − det A, so det A = 0 in this case.

15. Write d = det A, and let C denote the cofactor matrix of A. Here

AT = A−1 = 1
d adj A = d1 CT

Take transposes to get A = d1 C, whence C = dA.

 
0 c −c
19. b. Write A = −1 2 −1 . Then det A = c2 (Exercise 2) and the cofactor matrix is
c −c c

 2 −1 −1 −1 −1 2 
−
−c c c c c −c
   
c 0 −c
   c −c 0 −c 0 c

Ci j =  −
−c c c c
−
c −c
= 0 c2 c2
c c c
 
 
c −c 0 −c 0 c
−
2 −1 −1 −1 −1 2

   
 T c 0 c 1 0 1
1 1 1 1
Hence A−1 = det A adj A = c2
Ci j = c2
0 c2 c = c
0 c 1 for any c 6= 0.
−c c2 c −1 c 1
 3.2. Determinants and Matrix Inverses 47

 
4 c 3
d. Write A = c 2 c . Then det A = 2 (Exercise 2) and the cofactor matrix is
5 c 4

 2 c c c c 2 
−
c 4 5 4 5 c
  
8 − c2 c2 − 10

c
c 3 4 3 4 c
 
[Ci j ] =  −
c 4 5 4
−
5 c
= −c 1 c
c2 − 6 −c 8 − c2
 
 
c 3 4 3 4 c
−
2 c c c c 2

 T 1 8 − c2 −c c2 − 6
 
1 1
Hence A−1 = det A adj A = Ci j = 2 2 c 2
1 −c .
c − 10 c 8 − c2
 
1 c −1
f. Write A = c 1 1 . Then det A = −(c3 + 1) (Exercise 2) so det A = 0 means c 6= −1 (c
0 1 c
is real). The cofactor matrix is
 1 1 c 1 c 1 
−
1 c 0 c 0 1
−c2
   
c−1 c
   c −1 1 −1 1 c

Ci j =  −
1 c 0 c
−
0 1
= −(c2 + 1) c −1
c+1 −(1 + c) 1 − c2
 
 
c −1 1 −1 1 c
−
1 1 c 1 c 1

c − 1 −(c2 + 1) 1 − c c2 + 1 −c − 1
   
c+1
1 −1 −1 1
Hence A−1 = det A adj A = [C ]T
c3 +1 i j
= c3 +1
−c2 c −(c + 1) = c3 +1
c2 −c c+1 ,
c −1 1 − c2 −c 1 c2 − 1
where c 6= −1.

20. b. True. Write d = det A, so that d · A−1 = adj A. Since adj A = A−1 by hypothesis, this gives
dA−1 = A−1 , that is (d − 1)A−1 = 0. It follows that d = 1 because A−1 6= 0 (see Example
2.1.7).
d. True. Since AB = AC we get A(B − C) = 0. As A is invertible, this means B = C. More
precisely, left multiply by A−1 to get A−1 A(B − C) = A−1 0 = 0; that is I(B − C) = 0; that is
B −C = 0, so B = C.
 
1 1 1
f. False. If A = 1 1 1 then adj A = 0. However A 6= 0.
1 1 1
h i h i
1 1 0 −1
h. False. If A = 0 0 then adj A = 0 1 , and this has no row of zeros.
h i
−1 1
j. False. If A = 1 −1 then det (I + A) = −1 but 1 + det A = 1.
h i h i
l. False. If A = 10 11 then det A = 1, but adj A = 10 −11 6= A.

22. b. If p(x) = r0 + r1 x + r2 x2 , the conditions give linear equations for r0 , r1 and r2 :

r0 = p(0) = 5
r0 + r1 + r2 = p(1) = 3
r0 + 2r1 + 4r2 = p(2) = 5

The solution is r0 = 5, r1 = −4, r2 = 2, so p(x) = 5 − 4x + 2x2 .

48 Determinants and Diagonalization

23. b. If p(x) = r0 + r1 x + r2 x2 + r3 x3 , the conditions give linear equations for r0 , r1 , r2 and r3 :

r0 = p(0) = 1
r0 + r1 + r2 + r3 = p(1) = 1
r0 − r1 + r2 − r3 = p(−1) = 2
r0 − 2r1 + 4r2 − 8r3 = p(−2) = −3
−5
The solution is r0 = 1, r1 = 3 , r2 = 12 , r3 = 76 , so p(x) = 1 − 53 x + 12 x2 + 76 x3 .

24. b. If p(x) = r0 + r1 x + r2 x2 + r3 x3 , the conditions give linear equations for r0 , r1 , r2 and r3 :

r0 = p(0) = 1
r0 + r1 + r2 + r3 = p(1) = 1.49
r0 + 2r1 + 4r2 + 8r3 = p(2) = −0.42
r0 + 3r1 + 9r2 + 27r3 = p(3) = −11.33
The solution is r0 = 1, r1 = −0.51, r2 = 2.1, r3 = −1.1, so

p(x) = 1 − 0.51x + 2.1x2 − 1.1x3

The estimate for the value of y corresponding to x = 1.5 is

y = p(1.5) = 1 − 0.51(1.5) + 2.1(1.5)2 − 1.1(1.5)3 = 1.25

to two decimals.
26. b. Let A be an upper triangular, invertible, n × n matrix. We use induction
h i on n. If n h= 1 it isi
a X
clear (every 1 × 1 matrix is upper triangular). If n > 1 write A = 0 B and A−1 = Zb YC
in block form. Then h i h i
1 0 −1 ab + XZ aY + XC
0 I = AA = BZ BC

So BC = I, BZ = 0. Thus C = B−1 is upper triangular by induction (B is hupper itriangular

because A is) and BZ = 0 gives Z = 0 because B is invertible. Hence A−1 = b0 CY is upper
triangular.
 
3 0 1
28. Write d = det A. Then d = det (A ) = det 0 2 3 = −21. Hence d = −1
1 −1
21 . By Theorem 3.2.4,
3 1 −1
 
3 0 1
we have A · adj A = dI, so adj A = A−1 (dI) = dA−1 = −1
21
0 2 3 .
3 1 −1

34. b. Write d = det A so det A−1 = d1 . Now the adjugate for A−1 gives

A−1 ( adj A−1 ) = d1 I

Take inverses to get ( adj A−1 )−1 A = dI. But dI = ( adj A)A by the adjugate formula for A.
Hence
( adj A−1 )−1 A = ( adj A)A
−1
Since A is invertible, we get adj A−1

= adj A, and the result follows by taking inverses
again.
 3.3. Diagonalization and Eigenvalues 49

d. The adjugate formula gives

AB adj (AB) = det AB · I = det A · det B · I

On the other hand

AB adj B · adj A = A[( det B)I] adj A

= A · adj A · ( det B)I
= ( det A)I · ( det B)I
= det A det B · I

Thus AB adj (AB) = AB · adj B · adj A, and the result follows because AB is invertible.

3.3 Diagonalization and Eigenvalues

1. b. cA (x) = x −1 2 x +4 1 = x2 − x − 6 = (x − 3)(x + 2); hence the eigenvalues are λ1 = 3, and

λ2 = −2.h Take ithesehvaluesifor x inhthe matrix
i xI − A for cA (x):
1 4 1 4 4
λ1 = 3: 1 4 → 0 0 ; x1 = −1 .
h i h i h i
λ2 = −2: −41 −14 → 10 −10 ; x2 = 11 .
 h i h i
So P = x1 x2 = −14 11 has P−1 AP = 30 −20 .


d. To compute cA (x) we first add row 1 to row 3:

x−1 −1 3 x−1 −1 3 x−1 −1 −x + 4
cA (x) = −2 x −6 = −2 x −6 = −2 x −4
−1 1 x−5 x−2 0 x−2 x−2 0 0
−1 −x + 4
= (x − 2) x −4 = (x − 2)[x2 − 4x + 4] = (x − 2)3

So the eigenvalue is λ1 = 2 of multiplicity 3. Taking x = 3 in the matrix xI − A for cA (x):

         
1 −1 3 1 −1 3 s − 3t 1 −3
−2 2 6 → 0 0 0 , x= s ; x1 = 1 ; x2 = 0 .
−1 1 −3 0 0 0 t 0 1

Hence there are not n = 3 basic eigenvectors, so A is not diagonalizable.

x −1 0
f. Here cA (x) = −3 x −1 = x3 − 3x − 2. Note that −1 is a root of cA (x) so x + 1 is a
−2 0 x
factor. Long division gives cA (x) = (x + 1)(x2 − x − 2). But x2 − x − 2 = (x + 1)(x − 2), so
cA (x) = (x + 1)2 (x − 2). Hence, the eigenvalues are λ1 = −1 and λ2 = 2. Substitute λ1 = −1
in the matrix xI − cA (x) gives
  " 1
#
1 0
−1 −1 0 2
−3 −1 −1 → 0 1 − 12
−2 0 −1 0 0 0
50 Determinants and Diagonalization

so the solution involves only 1 parameter. As the multiplicity of λ1 is 2, A is not diagonaliz-

able by Theorem 3.3.5. Note that this matrix and the matrix in Example 3.3.9 have the same
characteristic polynomial but the matrix in Example 3.3.9 is diagonalizable, while this one is
not.
x−2 −1 −1
x−2 −1
h. cA (x) = 0 x−1 0 = (x − 1) −1 x−2 = (x − 1)2 (x − 3). Hence the eigenvalues are
−1 1 x−2
λ1 = 1, λ2 = 3. Take these  matrixxI − A for
 values forx in the (x):
 cA 
−1 −1 −1 1 1 1 1 0 1 −1
λ1 = 1: 0 0 0 → 0 2 0 → 0 1 0 ; x1 = 0
−1 1 −1 0 0 0 0 0 0 1
     
1 −1 −1 1 0 −1 1
λ2 = 3: 0 2 0 → 0 1 0 ; x2 = 0
−1 1 1 0 0 0 1
Since n = 3 and there are only two basic eigenvectors, A is not diagonalizable.
h i
2
2. b. As in Exercise 1, we find λ1 = 2 and λ2 = −1; with corresponding eigenvectors x1 = 1
h i h i h i
and x2 = 12 , so P = 21 12 satisfies P−1 AP = D = 20 −10 . Next compute
h i h ih i h i
2 −1 3 7
b= b1
b2 = P0−1 v = 1
3 −1 2 −1 = 1
3 −5

h i
Hence b1 = so, as λ1 is dominant, xk ∼
7
3 λ k x = 7 2k 2 .
= 1 1 1 3
b 1
     
1 1 1
d. Here λ1 = 3, λ2 = −2 and λ3 = 1; x1 = 0 , x2 = 1 and x3 = −2 , and P =
−3
    1   3
1 1 1 3 6 3 9
0 1 −2 . Now P−1 = 1 2 −2 −2 , so P−1 v0 = 1 2 and hence b1 = 23 . Hence
6 0 6
1 −3 1 −4 −1
3  1
1
vk ∼
= 32 3k 0 .
1

4. If λ is an eigenvalue for A, let Ax = λ x, x 6= 0. Then

A1 x = (A − α I)x = Ax − α x = λ x − α x = (λ − α )x
So λ − α is an eigenvalue of A1 = A − α I (with the same eigenvector). Conversely, if λ − α is
an eigenvalue of A1 , then A1 y = (λ − α )y for some y 6= 5. Thus, (A − α I)y = (λ − α )y, whence
Ay − α y = λ y − α y. Thus Ay = λ y so λ is an eigenvalue of A.
h i h in h i
8. b. Direct computation gives P−1 AP = 10 02 . Since 10 02 = 10 20n , the hint gives An =
h i h i
9 − 8 · 2n 12(1 − 2n )
P 10 20n P−1 = 6(2 n − 1) n
9·2 −8 .
h i
0 1
9. b. A = 0 . We have cA (x) = x(x − 2) so A has eigenvalues λ1 = 0 and λ2 = 2 with basic
2
h i h i  h 1 1 i
1 1 
eigenvectors x1 = 0 and x2 = 2 . Since x1 x2 = 0 2 is invertible, it is a diag-
h i
onalizing matrix for A. On the other hand, D + A = 10 11 is not diagonalizable by Example
3.3.10.
11. b. Since A is diagonalizable, let P−1 AP = D be diagonal. Then P−1 (kA)P = k(P−1 AP) = dD is
also diagonal, so kA is diagonalizable too.
 3.3. Diagonalization and Eigenvalues 51

d. Again let P−1 AP = D be diagonal. The matrix Q = U −1 P is invertible and

Q−1 (U −1 AU )Q = P−1U (U −1AU )U −1P = P−1 AP = D is diagonal.

This shows that U −1 AU is diagonalizable with diagonalizing matrix Q = U −1 P.

h i h i h i h i h i
1 1 2 1 −1 0 2 1 −1 0
12. 0 1 = 0 −1 + 0 2 and both 0 −1 and 0 2 are diagonalizable. However,
h i
1 1
0 1 is not diagonalizable by Example 3.3.10.

14. If A is n × n, let λ1 , λ2 , . . . , λn be the eigenvalues, all either 0 or 1. Since A is diagonalizable

(by hypothesis), we have P−1 AP = D where D = diag (λ1 , . . . , λn ) is the diagonal matrix with
λ1 , . . . , λn down the main diagonal. Since each λi = 0, 1 it follows that λi2 = λi for each i. Thus
D2 = diag (λ12 , . . . , λn2 ) = diag (λ1, . . . , λn ) = D. Since P−1 AP = D, we have A = PDP−1 . Hence

A2 = (PDP−1 )(PDP−1 ) = PD2 P−1 = PDP−1 = A

18. b. Since r 6= 0 and A is n × n, we have

x
] = rn det

x 
crA (x) = det [xI − rA] = det [r rI −A rI −A
x
As cA (x) = det [xI − A], this shows that crA (x) = rn cA


r .

20. b. If µ is an eigenvalue of A−1 then A−1 x = µ x for some column x 6= 0. Note that µ 6= 0 because
A−1 is invertible and x 6= 0. Left multiplication by A gives x = µ Ax, whence Ax = µ1 x. Thus,
1 1 1
µ is an eigenvalue of A; call it λ = µ . Hence, µ = λ as required. Conversely, if λ is any
eigenvalue of A then λ 6= 0 by (a) and we claim that λ1 is an eigenvalue of A−1 . We have
Ax = λ x for some column x 6= 0. Multiply on the left by A−1 to get x = λ A−1 x; whence
A−1 x = λ1 x. Thus λ1 is indeed an eigenvalue of A−1 .

21. b. We have Ax = λ x for some column x 6= 0. Hence, A2 x = λ Ax = λ 2 x, A3 = λ 2 Ax = λ 3 x, so

(A3 − 2A + 3I)x = A3 x − 2Ax + 3x = λ 3 x − 2λ x + 3x = (λ 3 − 2λ + 3)x

23. b. If λ is an eigenvalue of A, let Ax = λ x for some x 6= 0. Then A2 x = λ Ax = λ 2 x, A3 x = λ 2 Ax =

λ 3 x, . . . . We claim that Ak x = λ k x holds for every k ≥ 1. We have already checked this for
k = 1. If it holds for some k ≥ 1, then Ak x = λ k x, so

Ak+1 x = A(Ak x) = A(λ k x) = λ k Ax = λ k (λ x) = λ k+1 x

Hence, it also holds for k + 1, and so Ak x = λ k x for all k ≥ 1 by induction. In particular, if

Am = 0, m ≥ 1, then λ m x = Am x = 0x = 0. As x 6= 0, this implies that λ m = 0, so λ = 0.
24. a. Let A be diagonalizable with Am = I. If λ is any eigenvalue of A, say Ax = λ x for some column
x 6= 0, then (see the solution to 23(b) above) Ak x = λ k x for all k ≥ 1. Taking k = m we have
x = Am x = λ m x, whence λ m = 1. Thus λ is a complex mth root of unity and so lies on the unit
circle by Theorem A.3. But we are assuming that λ is a real number so λ = ±1, so λ 2 = 1.
Also A is diagonalizable, say P−1 AP = D = diag (λ1, . . . , λn ) where the λi are the eigenvalues
of A. Hence D2 = diag (λ12 , . . . , λn2 ) = I because λi2 = 1 for each i. Finally, since A = PDP−1
we obtain A2 = PD2 P−1 = PIP−1 = I.
52 Determinants and Diagonalization

27. a. If A is diagonalizable and has only one eigenvalue λ , then the diagonalization algorithm asserts
that P−1 AP = λ I. But then A = P(λ I)P−1 = λ I, as required.
b. Here the characteristic polynomial is cA (x) = (x − 1)2 , so the only eigenvalue is λ = 1. Hence
A is not diagonalizable by (a).
h 1 1 i h 1 1 i
31. b. The matrix in Example 3.3.1 is 22 04 . In this case A = 34 04 so
h i
x − 14 − 41
cA (x) = det −3 x = x2 − 14 x − 34 = (x − 1)(x + 34 )

Hence the dominant eigenvalue is λ = 1, and the population stabilizes.

h 3 1 i h i
− 35 − 51
d. In this case A = 35 05 so cA (x) = det x−3 x = x2 − 35 x − 35 . By the quadratic for-
1
√ 1
√
mula, the roots are 10 [3 ± 69], so the dominant eigenvalue is 10 [3 + 69] ≈ 1.13 > 1, so the
population diverges.
h 2
i
34. Here the matrix A in Example 3.3.1 is A = α2 05 where α is the adult reproduction rate. Hence
h i  q 
x − α − 25 2 4 1 2 16
cA (x) = det −2 x = x − α x − 5 , and the roots are 2 α ± α + 5 . Thus the dominant
 q 
1
eigenvalue is λ1 = 2 α + α + 5 , and this equals 1 if and only if α = 15 . So the population
2 16

stabilizes if α = 15 . In fact it is easy to see that the population becomes extinct (λ1 < 1) if and only
if α < 15 , and the population diverges (λ1 > 1) if and only if α > 15 .

3.4 An Application to Linear Recurrences

h i h ih i
0 1
1. b. In this case xk+2 = 2xk −xk+1 , so vk+1 = 2xkx−k+1xk+1 = 2 −1
xk
xk+1 = Avk . Diagonalizing
h i h i
1 1 1 0
A gives P = 1 −2 and D = 0 −2 . Hence
h i h ih i h 4
i
b= b1
b2 = P0−1 v0 = 1
3
2
1
1
−1
1
2 = 3
− 31
h i h i h i
xk 1 1
Thus xk+1 = 43 1k 1 − 13 (−2)k −2 for each k. Comparing top entries gives

(−2)k (−2)k
h i
1 k
xk = 34 − 13 (−2)k = 13 [4 − (−2)k ] =


3 1−4 −2 ≈− 3 for large k.

Here −2 is the dominant eigenvalue, so xk = 31 (−2)k [ (−2)

4 1 k
k − 1] ≈ − 3 (−2) if k is large.
h i h ih i
d. Here xk+2 = 6xk − xk+1 , so vk+1 = 6xkx−k+1xk+1 = 06 −11 xk
xk+1 = Avk .
h i h i
Diagonalizing A gives P = 12 −31 and D = 20 −30 . Hence
 
4
h i h ih i
b= b1
b2 = P0−1 v = 15 3
2
1
−1
1
1 = 5
1
5
 3.4. An Application to Linear Recurrences 53

h i h i h i
xk 1 1
Now xk+1 = 45 2k 2 + 15 (−3)k −3 , and so, looking at the top entries we get

xk = 54 2k + 15 (−3)k = 15 [2k+2 + (−3)k ]

(−3)k
i (−3)k
h
2 k
Here xk = 5 1 + r −3 ≈ 5 for large k, so −3 is dominant.
     
xk 0 1 0 1 1 1
2. b. Let vk = xk+1 . Then A = 0 0 1 , and diagonalization gives P = −1 −2 1 and
xk+2 2 1 −2 1 4 1
" #" # " 1 #
1 − 21 − 21 1
   
−1 0 0 b1 2
D= 0 −2 0 . Then b2 = P0−1 v0 = − 13 0 1
3 0 = 0 ,
0 0 1 b3 1 1 1
1 1
3 2 6 2
giving the general formula
     
1 1 1
1 k k 1 k
vk = 2 (−1)1
−1 + (0)(−2) −2 + 21
1
1 4 1

Thus equating first entries give

xk = 12 (−1)k + 12 1k = 12 [(−1)k + 1]

Note that the sequence xk here is 0, 1, 0, 1, 0, 1, . . . which does not converge to any fixed
value for large k.
3. b. If a bus is parked at one end of the row, the remaining spaces can be filled in xk ways to fill it in;
if a truck is at the end, there are xk+2 ways; and if a car is at the end, there are xk+3 ways. Since
one (and only one) of these three possibilities must occur, we have xk+4 = xk + xk+2 + xk+3
must hold for all k ≥ 1. Since x1 = 1, x2 = 2 (cc or t), x3 = 3 (ccc, ct or tc) and x4 = 6 (cccc,
cct, ctc, tcc, tt, b), we get successively, x5 = 10, x6 = 18, x7 = 31, x8 = 55, x9 = 96, x10 = 169.
5. Let xk denote the number of ways to form words of k letters. A word of k + 2 letters must end in
either a or b. The number of words that end in b is xk+1 — just add a b to a (k + 1)-letter word.
But the number ending in a is xk since the second-last letter must be a b (no adjacent a’s) so we
simply add ba to any k-letter word. This gives the recurrence xk+2 = xk+1 + xk which is the same as
in Example 3.4.2, but with different initial conditions: x0 = 1 (since the “empty” word is the only
one formed with no letters) and x1 = 2. The eigenvalues, eigenvectors, and diagonalization remain
the same, and so h i h i
vk = b1 λ1k λ11 + b2 λ2k λ12
√ √
where λ1 = 12 (1 + 5) and λ2 = 12 (1 − 5). Comparing top entries gives

xk = b1 λ1k + b2 λ2k
h i
By Theorem 2.4.1, the constants b1 and b2 come from bb12 = P0−1 v0 . However, we vary the method
and use the initial conditions to determine the values of b1 and b2 directly. More precisely, x0 = 1
means√ 1 = b1 + b2 while
√ x1 = 2 means 2 = b1 λ1 + b2 λ2 . These equations have unique solution
5−3 5−3
b1 = √ and b2 = √ . It follows that
2 5 2 5

√  1+√5 k √  1−√5 k
 
1
xk = √ (3 + 5) 2 + (−3 + 5) 2 for each k ≥ 0
2 5
54 Determinants and Diagonalization

7. In a stack of k + 2 chips, if the last chip is gold then (to avoid having two gold chips together)
the second last chip must be either red or blue. This can happen in 2xk ways. But there are xk+1
ways that the last chip is red (or blue) sohthere are
i 2xk+1 ways these possibilities can occur. Hence
0 1
√ √
xk+2 = 2xk + 2xk+1 . The matrix is A = 2 2 with eigenvalues λ1 = 1 + 3 and λ2 = 1 − 3
h i h i
1
and corresponding eigenvectors x1 = λ1 and x2 = λ12 . Given the initial conditions x0 = 1 and
x1 = 3, we get
h i h ih i h √ i h √ i
b1 −1 1 λ2 −1 1 1√ −2 −√ 3 1 2 + √3
b2 = P 0 v 0 = √
3 −λ1 1 3 = −2 3 2− 3
= √
2 3 −2 + 3

Since Theorem 2.4.1 gives h i h i

1 1
vk = b1 λ1k λ1 + b2 λ2k λ2

comparing top entries gives

1
h √ √ √ √ i
xk = b1 λ1k + b2 λ2k = √
2 3
(2 + 3)(1 + 3)k + (−2 + 3)(1 − 3)k

yk +yk+1
9. Let yk be the yield for year k. Then the yield for year k + 2 is yk+2 = =i12 yk + 12 yk+1
h . The
i 2 h
eigenvalues are λ1 = 1 and λ2 = − 12 , with corresponding eigenvectors x1 = 1 and x2 = −21 .
1

Given that k = 0 for the year 1990, we have the initial conditions y0 = 10 and y1 = 12. Thus
h i h ih i h i
b1 −1 1 1 2 10 1 34
b2 = P0 v 0 = 3 −1 1 12 = 3 2

Since h i
k h i
34 1 −2
vk = 3 (1)
k
1 + 23 − 12 1

then
k
k
34 2
yk = k
3 (1) + 3 (−2) − 12 = 34
3 − 43 − 21
34
For large k, yk ≈ so the long term yield is 11 13 million tons of wheat.
3
 
0 1 0
11. b. We have A = 0 0 1 so cA (x) = x3 − (a + bx + cx2 ). If λ is any eigenvalue of A, and we
 a b c
1
write x = λ2 , we have
λ

      
0 1 0 1 λ λ
Ax = 0 0 1 λ = λ2 = λ2 = λx
a b c λ2 a + bλ + cλ 2 λ3

because cA (λ ) = 0. Hence x is a λ -eigenvector.

5 5 11
12. h p=
b. We have i 6 from (a), so yk = xk + 6 satisfies yk+2 = yk+1 + 6yk withh y0 = y1i = 6 . Here
A = 06 11 with eigenvalues 3 and −2, and diagonalizing matrix P = 13 −12 . This gives
11
3 − (−2)k+1 , so xk = 11 k+1 − 5 .
 k+1   k+1 
yk = 30 30 3 − (−2) 6
 3.5. An Application to Systems of Differential Equations 55

13. a. If pk is a solution of (3.12) and qk is a solution of (3.13) then

qk+2 = aqk+1 + bqk

pk+2 = apk+1 + bpk + c(k)

for all k. Adding these equations we obtain

pk+2 + qk+2 = a(pk+1 + qk+1 ) + b(pk + qk ) + c(k)

that is pk + qk is also a solution of (3.12).

b. If rk is any solution of (3.12) then rk+2 = ark+1 + brk + c(k). Define qk = rk − pk for each k.
Then it suffices to show that qk is a solution of (3.13). But

qk+2 = rk+2 − pk+2 = (ark+1 + brk + c(k)) − (apk+1 + bpk + c(k)) = aqk+1 + bqk

which is what we wanted.

3.5 An Application to Systems of Differential Equations

h i
1. b. The matrix of the system is A = −11 53 so cA (x) = x−1
+1 −5
x − 3 = (x − 4)(x + 2).
h i h i h i
λ1 = 4: −15 −51 → 10 −10 ; an eigenvector is x1 = 11 .
h i h i h i
−1 −5 1 5 5
λ2 = −2: −1 −5 → 0 0 ; an eigenvector is x2 = −1 .
h i h i
Thus P−1 AP = 40 −20 where P = 11 −15 . The general solution is
h i h i
f = c1 x1 eλ1 x + c2 x2 eλ2 x = c1 1
1 e4x + c2 5
−1 e−2x

Hence, f1 (x) = c1 e4x + 5c2 e−2x , f2 (x) = c1 e4x − c2 e−2x . The boundary condition is f1 (0) = 1,
f2 (0) = −1; that is h i h i h i
1 1 5
−1 = f(0) = c1 1 + c2 −1

Thus c1 + 5c2 = 1, c1 − c2 = −1; the solution is c1 = − 23 , c2 = 13 , so the specific solution is

f1 (x) = 13 (5e−2x − 2e4x ), f2 (x) = − 31 (2e4x + e−2x )

 
2 1 2
d. Now A = 2 2 −2 . To evaluate cA (x), first subtract row 1 from row 3:
3 1 1

x−2 −1 −2 x−2 −1 −2 x−4 −1 −2

cA (x) = −2 x−2 2 = −2 x−2 2 = 0 x−2 2
−3 −1 x−1 −x − 1 0 x+1 0 0 x+1

= (x + 1)(x − 2)(x − 4)
56 Determinants and Diagonalization

8
       
−3 −1 −2 1
5 −6 1 0 7 −8
λ1 = −1: −2 −3 2 → 2
3 −2 → 0 1 − 10 7 ; x1 = 10
 −3 −1 −2  0
0 0
  0 0 0
  7

0 −1 −2 1 0 −1 1 0 −1 1
λ2 = 2: −2 0 2 → 0 1 2 → 0 1 2 ; x2 = −2
 −3 −1 1   0 −1 −2
 0 0 0
  1 
2 −1 −2 2 −1 −2 1 0 −1 1
λ3 = 4: −2 2 2 → 0 2 0 → 0 1 0 ; x3 = 0
−3 −1 −3 −1
 3  3
 0 0
 0 1
−1 0 0 −8 1 1
Thus P−1 AP = 0 2 0 where P = 10 −2 0 . The general solution is
0 0 4 7 1 1

     
−8 1 1
−x 2x 4x −x 2x
f = c1 x1 e + c2 x2 e + c3 x3 e = c1 10 e + c2 −2 e + c3 0 e4x
7 1 1

That is

f1 (x) = −8c1 e−x + c2 e2x + c3 e4x

f2 (x) = 10c1 e−x − 2c2 e2x
f3 (x) = 7c1 e−x + c2 e2x + c3 e4x

If we insist on the boundary conditions f1 (0) = f2 (0) = f3 (0) = 1, we get

−8c1 + c2 + c3 = 1
10c1 − 2c2 = 1
2c1 + c2 + c3 = 1

The coefficient matrix is P is invertible, so the solution is unique: c1 = 0, c2 = − 12 , c3 = 23 .

Hence

f1 (x) = 12 (3e4x − e2x )

f2 (x) = e2x
f3 (x) = 12 (3e4x − e2x )

Note that f1 (x) = f3 (x) happens to hold.

3. b. Have m′ (t) = km(t), so m(t) = cekt by Theorem 3.5.1. Then the requirement that m(0) = 10
gives c = 10. Also we ask that m(3) = 8, whence 10e3k = 8, e3k = 45 . Hence (ek )3 = 45 , so
(ek ) = ( 45 )1/3 . Thus m(t) = 10( 45 )t/3 . Now, we want the half-life t0 satisfying m(t0) = 12 m(0),
3 ln(1/2)
that is 10( 45 )t0 /3 = 5 so t0 = ln(4/5) = 9.32 hours.

5. a. Assume that a g′ = Ag where A is n × n. Put f = g − A−1 b where b is a column of constant

functions. Then f′ = g′ = Ag = A(f + A−1 b) = Af + b, as required.

6. b. Assume that f1′ = a1 f1 + f2 and f2′ = a2 f1 . Differentiating gives f1′′ = a1 f1′ + f2′ = a1 f1′ + a2 f1 .
This shows that f1 satisfies (∗).
 3.6. Proof of the Cofactor Expansion Theorem 57

3.6 Proof of the Cofactor Expansion Theorem

2. Consider the rows R p , R p+1 , . . . , Rq−1 , Rq . Using adjacent interchanges we have

     
Rp R p+1 Rq
 R p+1  −−−−−−−−−→  ..  −−−−−−−−−→  R p+1 
 . 
 .. 
   
   .. 
 .   Rq−1   . 

 Rq−1 
 q− p  Rq  q − p − 1  Rq−1 
   
interchanges interchanges
Rq Rp Rp

Hence 2(q − p) − 1 interchanges are used in all.

Supplementary Exercises: Chapter 3

2. b. Proceed by induction on n where A is n × n. If n = 1, AT = A. In general, induction and (a)

give

T
det [Ai j ] = det [ Ai j ] = det [(AT )i j ]
Write AT = [a′i j ] where a′i j = a ji , and expand det (AT ) along column 1:
n n
det (AT ) = ∑ a′j1(−1) j+1 det [(AT ) j1] = ∑ a1 j (−1)1+ j det [A1 j ] = det A
j=1 j=1

where the last equality is the expansion of det A along row 1.

4. Vector Geometry

4.1 Vectors and Lines

√
 
1 p
1. b. −1 = 12 + (−1)2 + 22 = 6
2

√
 
−1 p
d. 0 = (−1)2 + 02 + 22 = 5
2

√ √
 
1
f. −3 1 = |−3| 12 + 12 + 22 = 3 6
2
   
−2 −2
2. b. A vector u in the direction of must have the form u = t
−1 for a scalar t > 0. Since
−1
2 2
q
u is a unit vector, we want kuk = 1; that is 1 = |t| (−2)2 + (−1)2 + 22 = 3t, which gives
 
−2
1 1
t = 3 . Hence u = 3 −1 .
2
   
2 2
4. b. Write u = and v =
−1 0 . The distance between u and v is the length of their difference:
2 1
√
 
0 p
ku − vk = −1 = 02 + (−1)2 + 12 = 2.
1
     
4 3 1 p
d. As in (b), the distance is 0 − 2 = −2 = 12 + (−2)2 + (−2)2 = 3.
−2 0 −2

−→ −
→
6. b. In the diagram, let E and F be the midpoints of sides BC and AC respectively. Then FC = 12 AC
−→ −
→
and CE = 12 CB. Hence
−→ −→ −→ 1 − → −
→ −
→ − → −
→
FE = FC + CE = 2 AC + 12 CB = 12 (AC + CB) = 12 AB

7. Two nonzero vectors are parallel if and only if one is a scalar multiple of the other.

b. Yes, they are parallel: u = (−3)v.

d. Yes, they are parallel: v = (−4)u.
−→
8. b. QR = p because OPQR is a parallelogram (where O is the origin).
−→ −→
d. RO = −(p + q) because OR = p + q.

59
60 Vector Geometry

√ √
     
−→ 1 2 −1 −→ p
9. b. PQ = −1 − 1 = −1 , so PQ = (−1)2 + (−1)2 + 52 = 27 = 3 3.
6 0 5
−→ −→
d. Here P = Q are equal points, so PQ = 0. Hence PQ = 0.
√
       
−→ 1 3 −2 −1 −→ p
f. PQ = 1 − −1 = 2 =2 1 . Hence PQ = |2| (−1)2 + 12 + (−1)2 = 2 3.
4 6 −2 −1
   
x 3 −→
10. b. Given Q(x, y, z) let q = y and p = 0 be the vectors of Q and P. Then PQ = q − p.
z −1
 
2
Let v = −1 .
3
 
−→ 5
(i) If PQ = v then q − p = v, so q = p + v = . Thus Q = Q(5, −1, 2). −1
2
 
−→ 1
(ii) If PQ = −v then q − p = −v, so q = p − v = 1 . Thus Q = Q(1, 1, −4).
−4

11. b. If 2(3v − x) = 5w + u − 3x then 6v − 2x = 5w + u − 3x, so

       
−5 3 24 −26
x = 5w + u − 6v = 5 + −1 − 0 = 4
25 0 6 19

       
a 0 c a+c
12. b. We have au + bv + cw = + + a b 0 = a+b . Hence setting
2a 2b −c 2a + 2b − c
 
1
au + bv + cw = x = 3 gives equations
0

a + c = 1
a + b = 3
2a + 2b − c = 0

The solution is a = −5, b = 8, c = 6.

   
5 3a + 4b + c
13. b. Suppose 6 = au + bv + cw = −a + c . Equating coefficients gives linear equations
−1 b+c
for a, b, c:
3a + 4b + c = 5
−a + c = 6
b + c = −1
This system has no solution, so no such a, b, c exist.
     
x 2 1
14. b. Write P = P(x, y, z) and let p = y , p1 = 1 and p2 = −2 be the vectors of P, P1
z −2 0
and P2 respectively. Then
−→ −−→
p − p2 = P2 P = p2 + 14 (P2 P1) = p2 + 14 (p1 − p2 ) = 41 p1 + 34 p2
 4.1. Vectors and Lines 61

Since p1 and p2 are known, this gives

     
2 1 5
1 3 1
p= 4 1 + 4
−2 = 4
−5
−2 0 −2

5
− 54 , − 12 .


Hence P = P 4,
−→ −→
17. b. Let p =
 OP and q = OQ
 denote
 the vectors of the points
  and
P    Then q − p =
 Q respectively.
−→ −1 1 −1 1 0
PQ = 4 and p = 3 , so q = (q − p) + p = 4 + 3 = 7 .
7 −4 7 −4 3
Hence Q = Q(0, 7, 3).

18. b. We have kuk2 = 20, so the given equation is 3u + 7v = 20(2x + v). Solving for x gives
     
6 26 −20
40x = 3u − 13v = 0 − 13 = −13
−12 −26 14

 
−20
1
Hence x = 40
−13 .
14

−
→ −→
20. b. Let S denote the fourth point. We have RS = PQ, so
     
−
→ −→ − → −→ −→ 3 −4 −1
OS = OR + RS = OR + PQ = −1 + 4 = 3
0 2 2

Hence S = S(−1, 3, 2).

21. b. True. If kv − wk = 0 then v − w = 0 by Theorem 4.1.1, so v = w.

d. False. kvk = k−vk for all v but v = −v only holds if v = 0.
f. False. If t < 0 they have opposite directions.
h. False. By Theorem 4.1.1, k−5vk = |−5| kvk = 5 kvk so it fails if v 6= 0.
j. False. If w = −v where v 6= 0, then kv + wk = 0 but kvk + kwk = 2 kvk =
6 0.
   
−→ 2 3
22. b. One direction vector is d = QP = −1 . Let p0 = −1 be the vector of P. Then the vector
5 4
equation of the line is
     
3 2 x
p = p0 + td = −1 +t −1 when p = y
4 5 z

is the vector of an arbitrary point on the line. Equating coefficients gives the parametric equa-
tions of the line

x = 3 + 2t
y = −1 − t
z = 4 + 5t
62 Vector Geometry

   
1 1
d. Now p0 = 1 because P1 (1, 1, 1) is on the line, and take d = because the line is to 1
1 1
     
1 1 x
be parallel to d. Hence the vector equation is p = p0 + td = 1 + t 1 . Taking p = y ,
1 1 z
x = 1+t
the scalar equations are y = 1 + t .
z = 1+t
x = 2−t 
−1

f. The line with parametric equations y = 1 has direction vector d = 0 — the compo-
1
z=t
nents are the coefficients of t. Sinceour line is parallel to this one, d will do as direction vector.
2
We are given the vector p0 = −1 of a point on the line, so the vector equation is
1

   
2 −1
p = p0 + td = −1 +t 0
1 1

The scalar equations are

x = 2−t
y = −1
z = 1+t
   
x 4−t
23. b. P(2, 3, −3) lies on the line y = 3 since it corresponds to t = 2. Similarly Q(−1, 3, −9)
z 1 − 2t
corresponds to t = 5, so Q lies on the line too.

24. b. If P = P(x, y, z) is a point on both lines then

x = 1−t
y = 2 + 2t for some t because P lies on the first line.
z = −1 + 3t

x = 2s
y = 1 + s for some s because P lies on the second line.
z=3
If we eliminate x, y, and z we get three equations for s and t:

1 − t = 2s
2 + 2t = 1 + s
−1 + 3t = 3

The last two equations require t = 43 and s = 11

3 , but these values do not satisfy the first equation.
Hence no such s and t exist, so the lines do not intersect.
 4.1. Vectors and Lines 63

 
x
d. If y is the vector of a point on both lines, then
z

     
x 4 1
y = −1 +t 0 for some t (first line)
z 5 1
     
x 2 0
y = −7 +s −2 for some s (second line).
z 12 3

         
x 4 1 2 0
Eliminating y gives −1 +t 0 = −7 +s −2 . Equating coefficients gives three
z 5 1 12 3
equations for s and t:

4+t = 2
−1 = −7 − 2s
5 + t = 12 + 3s

This has a (unique) solution t = −2, s = −3 so the lines do intersect. The point of intersection
has vector          
4 1 4 1 2
−1 +t 0 = −1 −2 0 = −1
5 1 5 1 3
         
2 0 2 0 2
(equivalently −7 +s −2 = −7 −3 −2 = −1 ).
12 3 12 3 3
     
1 2 1
29. Let a = −1 and b = 0 be the vectors of A and B. Then d = b − a = 1 is a direction
2 1 −1
vector for the line through A and B, so the vector c of C is given by c = a + td for some t. Then
−
→ −
→
AC = kc − ak = ktdk = |t| kdk and BC = kc − bk = k(t − 1)dk = |t − 1| kdk

−
→ −→
Hence AC = 2 BC means |t| = 2 |t − 1|, so t 2 = 4(t −1)2, whence 0 = 3t 2 −8t +4 = (t −2)(3t −
  " 5 #
3 3
2
2). Thus t = 2 or t = 3 . Since c = a + td, this means c = 1 or c = − 13 .
0 4
3

31. b. If there are 2n points, then Pk and Pn+k are opposite ends of a diameter of the circle for each
−
→ −→ −
→ − → −
→
k = 1, 2, . . . . Hence CPk = −CPn+k so these terms cancel in the sum CP1 + CP2 + · · · + CP2n .
Thus all terms cancel and the sum is 0.
−→ −→ −→ −→
33. We have 2EA = DA because E is the midpoint of side AD, and 2AF = FC because F is 31 the way
−→ − →
from A to C. Finally DA = CB because ABCD is a parallelogram. Thus
−→ −→ −−→ −→ −→ −→ −→ − → −→ −→
2EF = 2(EA + AF) = 2EA + 2AF = DA + FC = CB + FC = FB
−→ −→
Hence EF = 12 FB so F is in the line segment EB, 1
3 the way from E to B. Hence F is the trisection
point of both AC and EB.
64 Vector Geometry

4.2 Projections and Planes

1. b. u · v = u · u = 12 + 22 + (−1)2 = 6
d. u · v = 3 · 6 + (−1)(−7) + 5(−5) = 18 + 7 − 25 = 0
f. v = 0 so u · v = a · 0 + b · 0 + c · 0 = 0
−18−2+0
2. b. cos θ = u·v
kukkvk = √ √
10 40
= −20
20 = −1. Hence θ = π.
d. cos θ = u·v
= √ = 1 . Hence θ = π .
√6+6−3
kukkvk 6(3 6) 2 3
0−21−4 2π
f. cos θ = u·v
kukkvk = √ √ = − 12 . Hence θ = 3 .
25 100
   
2 1
3. b. Writing u = −1 and v = x , the requirement is
1 2

1
2 = cos π3 = u·v
kukkvk = √2−x+2
√
6 x2 +5

Hence 6(x2 + 5) = 4(4 − x)2 , whence x2 + 16x − 17 = 0. The roots are x = −17 and x = 1.

4. b. The conditions are u1 · v = 0 and u2 · v = 0, yielding equations

3x − y + 2z = 0
2x + z = 0
 
−1
The solutions are x = −t, y = t, z = 2t, so v = t 1 .
2

d. The conditions are u1 · v = 0 and u2 · v = 0, yielding equations

2x − y + 3z = 0
0=0
   
1 0
The solutions are x = s, y = 2s + 3t, z = t, so v = s 2 + t 3 .
0 1

  2
−→ 2 3
6. b. PQ = −2 = 9 + 4 + 16 = 29
4
  2
−→ 2 2
QR = 7 = 4 + 49 + 4 = 57
2
2
  2
−
→ 5
PR = 5 = 25 + 25 + 36 = 86
6
−→ −→ 2 −→ 2
Hence PR = PQ + QR . Note that this implies that the triangle is right angled, that
PR is the hypotenuse, and hence that the angle at Q is a right angle. Of course, we can confirm
−→ −→
this latter fact by computing PQ · QR = 6 − 14 + 8 = 0.
 4.2. Projections and Planes 65

   
−
→ 2 −
→ 1
8. b. We have AB = 1 and AC = 2 so the angle α at A is given by
1 −1

−
→− →
AB·AC 2+2−1 1
cos α = −
→ − → = √ √
6 6
= 2
AB AC

   
π −
→ −2 −
→ −1
Hence α = 3 or 60◦ . Next BA = −1 and BC = 1 so the angle β at B is given by
−1 −2

−
→− →
BA·BC 2−1+2 1
cos β = −
→ − → = √ √ = 2
BA BC 6 6

Hence β = π3 . Since

π π π
the angles in any triangle add to π , the angle γ at C is π − 3 − 3 = 3 .
−
→ −1 −
→ 1
However, CA = −2 and CB = −1 , this can also be seen directly from
1 2

−
→− →
CA·CB −1+2+2 1
cos γ = −
→ − → = √ √ = 2
CA CB 6 6

   
4 4
u·v 12−2+1 11
10. b. proj v u = v = 16+1+1 1 = 1
kvk2 1
18 1
     
−6 −6 3
d. proj v u = u·v
v = −18−8−2 4 = − 12 4 = −2
kvk2 36+16+4 2 2 −1
       
−2 −2 3 −2
11. b. Take u1 = proj v u = u·v
v = −6+1+0
= 1 −5 1= . Then u2 = u−u1 = 1 5
+ 21 1
kvk2 4+1+16 4
21 4 0 4
 
53
1 26 . As a check, verify that u2 · v = 0, that is u2 is orthogonal to v.
21 20
   
−6 6
u·v −18−8−1 27
d. Take u1 = proj v u = 2 v = 36+16+1 4 = 53 −4 . Then u2 is given by u2 = u − u1 =
kvk −1 1
     
3 6 −3
27 1
−2 − 53 −4 = 53 2 . As a check, verify that u2 · v = 0, that is u2 is orthogonal to v.
1 1 26
      

1 3 −→ 1 0
12. b. Write p0 = 0 , d= 1 , p=
and write u = P0 P = p − p0 = −1 . Write
−1
−1 4 3 4
   
−−→ −→ 0−1+16 3 3
u1 = P0 Q and compute it as u1 = proj d P0 P = 9+1+16 1 = 15 26
1 . Then the distance
4 4
√
 
−→ 1
−45
1
from P to the line is QP = ku − u1 k = 26 −41 = 26 5642. To compute Q let q be its
44
vector. Then      
1 3 71
15 1
q = p0 + u1 = 0 + 26
1 = 26
15
−1 4 34

Hence Q = Q( 71 15 34
26 , 26 , 26 ).
   
i 3 −6 0
13. b. u × v = det j −1 2 = 0i − 0j + 0k = 0 =0
k 0 0 0
66 Vector Geometry

   
i 2 1 4
d. u × v = det j 0 4 = 4i − 15j + 8k = −15
k −1 7 8
       
−
→ − → −1 3 i −1 3 23
14. b. A normal is n = AB × AC = 1 × 8 = det j 1 8 = −32 .
−5 −17 k −5 −17 −11
Since the plane passes through B(0, 0, 1) the equation is

23(x − 0) − 32(y − 0) − 11(z − 1) = 0, that is − 23x + 32y + 11z = 11

 
2
d. The plane with equation 2x − y + z = 3 has normal n = −1 . Since our plane is parallel
1
to this one, n will serve as normal. The point P(3, 0, −1) lies on our plane, the equation is
2(x − 3) − (y − 0) + (z − (−1) = 0, that is 2x − y + z = 5.
 
−→ 1
f. The plane contains P(2, 1, 0) and P0 (3, −1, 2), so the vector u = PP0 = −2 is parallel
2
 
1
to the plane. Also the direction vector d = 0 of the line is parallel to the plane. Hence
−1
   
i 1 1 2
n = u × d = det j −2 0 = 3 is perpendicular to the plane and so serves as a normal.
k 2 −1 2
As P(2, 1, 0) is in the plane, the equation is

2(x − 2) + 3(y − 1) + 2(z − 0) = 0, that is 2x + 3y + 2z = 7

   
1 2
h. The two direction vectors d1 = −1 and d2 = 1 are parallel to the plane, so
−1
  3 
i 1 2 −2
n = d1 × d2 = det j −1 1 = 7 will serve as normal. The plane contains P(3, 1, 0)
k 3 −1 3
so the equation is

−2(x − 3) + 7(y − 1) + 3(z − 0) = 0, that is − 2x + 7y + 3z = 1

     
x 3 1
Note that this plane contains the line a = 1 + t −1 by construction; it contains the
z 0 3
other line because it contains P(0, −2, 5) and is parallel to d2 . This implies that the lines
intersect (both are in the same plane). In fact the point of intersection is P(4, 0, 3) [t = 1 on
the first line and t = 2 on the second line].
j. The set of all points R(x, y, z) equidistant from both P(0, 1, −1) and Q(2, −1, −3) is deter-
−→ −→ → 2
− −→ 2
mined as follows: The condition is PR = QR , that is PR = QR , that is

x2 + (y − 1)2 + (z + 1)2 = (x − 2)2 + (y + 1)2 + (z + 3)2

This simplifies to x2 + y2 + z2 − 2y + 2z + 2 = x2 + y2 + z2 − 4x + 2y + 6z + 14; that is 4x − 4y −

4z = 12; that is x − y − z = 3.
 
2
15. b. The normal n = 1 to the given plane will serve as direction vector for the line. Since the
0
     
x 2 2
line passes through P(2, −1, 3), the vector equation is y = −1 + t 1 .
z 3 0
 4.2. Projections and Planes 67

   
1 1
d. The given lines have direction vectors d1 = 1 and d2 = 2 , so
−2 −3
   
i 1 1 1
d = d1 × d2 = det j 1 2 = 1 is perpendicular to both lines.
k −2 −3 1

Hence d is a direction vector for the line we seek. As P(1, 1, −1) is on the line, the vector
equation is      
x 1 1
y = 1 +t 1
z −1 1
 
−→ 1+t
f. Each point on the given line has the form Q(2 + t, 1 + t, t) for some t. So PQ = t . This
t −2
 
−→ 1
is perpendicular to the given line if PQ · d = 0 (where d = 1 is the direction vector of the
1
given line). This condition is (1 + t) + t + (t − 2) = 0, that is t = 31 . Hence the line we want
" 4 #  
3 4
has direction vector 1
3 . For convenience we use d = 1 . As the line we want passes
− 35 −5
     
x 1 4
1 . [Note that Q 7 , 4 , 1 is


through P(1, 1, 2), the vector equation is y = 1 + t 3 3 3
z 2 −5
the point of intersection of the two lines.]
 
−→ 3
16. b. Choose a point P0 in the plane, say P0 (0, 6, 0), and write u = P0 P = −5 . Now write
−1
 
2
n= 1 for the normal to the plane. Compute
−1
 
2
u·n 2
u1 = proj n u = n = 1
knk2 6 −1
√
The distance from P to the plane is ku1 k = 13 6.
 
0
Since p0 = 6 and q are the vectors of P0 and Q, we get
0
       
0 3 2 7
1 1
q = p0 + (u − u1 ) = 6 + −5 − 3
1 = 3
2
0 −1 −1 −2

7 2 −2


Hence Q = Q 3, 3, 3 .
17. b. A normal to the plane is given by
       
−→ − → −2 −3 i −2 −3 −10
n = PQ × PR = 2 × −1 = det j 2 −1 = 6
−4 −3 k −4 −3 8

Thus, as P(4, 0, 5) is in the plane, the equation is

−10(x − 4) + 6(y − 0) + 8(z − 5) = 0; that is 5x − 3y − 4z = 0.

The plane contains the origin P(0, 0, 0).

68 Vector Geometry

19. b. The coordinates of points of intersection satisfy both equations:

3x + y − 2z = 1
x+y+z = 5

Solve
− 23
h i h i h i
3 1 −2 1 1 1 1 5 1 0 −2
1 1 1 5 → 0 −2 −5 −14 → 0 1 5
7
2

Take z = 2t, to eliminate fractions, whence x = −2 + 3t and y = 7 − 5t. Thus

       
x −2 + 3t −2 3
y = 7 − 5t = 7 = +t −5
z 2t 0 2

is the line of intersection.

20. b. If P(x, y, z) is an intersection point, then x = 1 + 2t, y = −2 + 5t, z = 3 − t since P is on the

line. Substitution in the equation of the plane gives 2(1 + 2t) − (−2 + 5t) − (3 − t) = 5, that is
1 = 5. Thus there is no such t, so the line does not intersect the plane.
d. If P(x, y, z) is an intersection point, then x = 1 + 2t, y = −2 + 5t and z = 3 −t since P is on the
line. Substitution in the equation of the plane gives −1(1 + 2t) − 4(−2 + 5t) − 3(3 − t) = 6,
  " 3 #
x 19
whence t = −8 3
, − 78 65


19 . Thus
y = − 1978 so P 19 19 , 19 is the point of intersection.
z 65
19

 
3
21. b. The line has direction vector d = 0 which is a normal to all such planes. If P0 (x0 , y0 , z0 )
2
is any point, the plane 3(x − x0 ) = 0(y − y0 ) + 2(z − z0) = 0 is perpendicular to the line. This
can be written 3x + 2z = 3x0 + 2z0 , so 3x + 2z = d, d arbitrary.
 
a
d. If the normal is n = b 6= 0, the plane is a(x − 3) + b(y − 2) + c(z + 4) = 0, where a, b and
c
c are not all zero.
   
−→ −1 a
f. The vector u = PQ = is parallel to these planes so the normal n =
1 is orthogonal b
−1 c
 
a
to u. Thus 0 = u · n = −a + b − c. Hence c = b − a and n = b . The plane passes through
b−a
Q(1, 0, 0) so the equation is a(x−1)+b(y−0)+(b−a)(z−0) = 0, that is ax+by+(b−a)z =
a. Here a and b are not both zero (as n 6= 0). As a check, observe that this plane contains
P(2, −1, 1) and Q(1, 0, 0).
 
a
h. Such a plane contains P0 (3, 0, 2) and its normal n = b must be orthogonal to the direction
c
   
1 a
vector d = −2 of the line. Thus 0 = d · n = a − 2b − c, whence c = a − 2b and n = b
−1 a − 2b
(where a and b are not both zero as n 6= 0). Thus the equation is

a(x − 3) + b(y − 0) + (a − 2b)(z − 2) = 0, that is ax + by + (a − 2b)z = 5a − 4b

where a and b are not both zero. As a check, observe that the plane contains every point
P(3 + t, −2t, 2 − t) on the line.
 4.2. Projections and Planes 69

23. b. Choose P1 (3, 0, 2) on the first line. The distance in question is the distance
 from P1 to
 the
−−→ 4 3
second line. Choose P2 (−1, 2, 2) on the second line and let u = P2 P1 = −2 . If d = 1
0 0
is the direction vector for the line, compute
u·d 10
 T  T
u1 = proj d u = d = 3 1 0 = 3 1 0
kdk2 10

 T √
then the required distance is ku − u1 k = 3 1 0 = 10.
     
1 3 −1
24. b. The cross product n = 1 × 1 =
of the two direction vectors serves as a normal to
3
1 0 −2
 
−−→ 1
the plane. Given P1 (1, −1, 0) and P2 (−2, −1, 3) on the lines, let u = P1 P2 = 0 . Compute
3

   
−1 1
−7 1
u1 = proj n u = 14
3 = 2
−3
−2 2

√ √
The required distance is ku1 k = 21 1 + 9 + 4 = 12 14.
Now let A = A(1 + s, −1 + s, s) and B = B(2 +  3t, −1 + t, 3) be the points on the two lines
−
→ 1 + 3t − s
that are closest together. Then AB = t −s is orthogonal to both direction vectors d1 =
3−s
   
1 3 −
→ −→
1 and d2 = 1 . By Theorem 4.2.3 this means d1 · AB = 0 = d2 · AB, giving equations
1 0
4t − 3s = −4, 10t − 4s = −3. The solution is t = 12 , s = 2, so the points are A = A(3, 1, 2) and
B = B 72 , − 12 , 3 .


√
d. Analogous to (b). The distance is 66 , and the points are A( 19 1 37 13


3 , 2, 3 ) and B = B 6 , 6 , 0 .

26. b. Position the cube with one vertex at the origin and sides along the  positive
 axes. Assume
a
each side has length a and consider the diagonal with direction d = a . The face diagonals
a
     
a a 0
that do not meet d are: ± −a , ± 0 and ± a , and all are orthogonal to d (the dot
0 −a −a
product is 0).

28. Position the solid with one vertex at the origin

 and sides,
 of lengths
 a,  b, c, along
 the positive x, y
a −a a a
and z axes respectively. The diagonals are ± b , ± b ,± −b and ± b . The possible
c c c −c
dot products are ±(−a2 + b2 + c2 ), ±(a2 − b2 + c2 ), ±(a2 + b2 − c2 ) and one of these is zero if and
only if the sum of two of a2 , b2 and c2 equals the third.

34. b. The sum of the squares of the lengths of the diagonals equals the sum of the squares of the
lengths of the four sides.

38. b. The angle θ between u and u + v + w is given by

u·(u+v+w) u·u+u·v+u·w kuk2 +0+0 kuk

cos θ = kukku+v+wk = kukku+v+wk = kukku+v+wk = ku+v+wk
70 Vector Geometry

Similarly the angles ϕ , ψ between v and w and u + v + w are given by

kvk kwk
cos ϕ = ku+v+wk and cos ψ = ku+v+wk

Since kuk = kvk = kwk qwe get cos θ = cos ϕ = cos ψ√

, whence θ = ϕ = ψ .
NOTE: ku + v + wk = kuk2 + kvk2 + kwk2 = kuk 3 by part (a), so cos θ = cos ϕ = cos ψ =
√1 . Thus, in fact θ = ϕ = ψ = .955 radians, (54.7◦ ).
3
−−→ h i
39. b. If P1 (x, y) is on the line then ax + by + c = 0. Hence u = P1 P0 = yx00 − x1
− y1 so the distance is

u·n |u·n| |a(x0 −x)+b(y 0 −y)| |ax√

0 +by0 +c|
k proj n uk = n = = √ =
knk2 knk a2 +b2 a2 +b2

41. b. This follows from (a) because kvk2 = a2 + b2 + c2 .

44. d. Take x1 = z2 = x, y1 = x2 = y and z1 = y2 = z in (c).

4.3 More on the Cross Product

   
i 1 3 5
3. b. One vector orthogonal to u and v is u × v = det = j . We have ku × vk =
2 1 −5
k −1 2 −5
√ √
     
1 5 1
1 3
5 −1 = 5 3. Hence the unit vectors parallel to u × v are ± 5√3 −5 = ± 3 −1 .
−1 −5 −1

1
4. b. The area of the triangle is the area of the parallelogram ABCD. By Theorem 4.3.4,
2
     
1 − → − → 2 4 0
1 1
Area of triangle = 2 AB × AC = 2 1 × 2 =2 0 =0
−1 −2 0

−
→ −
→
Hence AB and AC are parallel.
√
d. Analogous to (b). Area = 5.
 
−4
5. b. We have u × v = 5 so w · (u × v) = −7. The volume is |w · (u × v)| = |−7| = 7 by
1
Theorem 4.3.5.
6. b. The line through P0 perpendicular to the plane has direction vector n, and so has vector
 T
equation p = p0 + tn where p = x y z . If P(x, y, z) also lies in the plane, then
n · p = ax + by + cz = d. Using p = p0 + tn we find

d = n · p = n · p0 + t(n · n) = n · p0 + t knk2
 
d−n·p0 d−n·p0
Hence t = 2 , so p = p0 + 2 n. Finally, the distance from P0 to the plane is
knk knk

−→ 
d−n·p0

|d−n·p0 |
PP0 = kp − p0 k = 2 n = knk
knk
 4.4. Linear Operators on R3 71

10. The points A, B and C are all on one line if and only if the parallelogram they determine has area
−
→ − → −
→ − →
zero. Since this area is kAB × ACk, this happens if and only if AB × AC = 0.

12. If u and v are perpendicular, Theorem 4.3.4 shows that ku × vk = kuk kvk. Moreover, if w is
perpendicular to both u and v, it is parallel to u × v so w · (u × v) = ± kwk ku × vk because the angle
between them is either 0 or π . Finally, the rectangular parallepiped has volume

|w · (u × v)| = kwk ku × vk = kwk (kuk kvk)

using Theorem 4.3.5.

     
x p l
15. b. If u = y , v = q and w = m then, by the row version of Exercise 3.1.19 Section 3.1,
z r n
we get
 
i x l+p
u × (v + w) = det j y m+q
k z n+r
   
i x p i x l
= det j y q + det j y m = u×v+u×w
k z r k z n

     
v1 w1 u1
16. b. Let v = v2 ,w= w2 and u = u2 . Compute
v3 w3 u3

v · [(u × v) + (v × w) + (w × u)] = v · (u × v) + v · (v × w) + v · (w × u)
 
v1 w1 u1
= 0 + 0 + det v2 w2 u2
v3 w3 u3

by Theorem 4.3.1. Similarly

 
w1 u1 v1
w · [[(u × v) + (v × w) + (w × u)]] = w · (u × v) = det w2 u2 v2
w3 u3 v3

These determinants are equal because each can be obtained from the other by two column
interchanges. The result follows because (v − w) · x = v · x − w · x for any vector x.

22. If v1 and v2 are vectors of points in the planes (so v1 · n = d1 and v2 · n = d2 ), the distance is the
length of the projection of v2 − v1 along n; that is
 
(v2 −v1 )·n |(v −v )·n| |d −d |
k proj n (v2 − v1 )k = knk2
n = 2 knk1 = 2knk 1

4.4 Linear Operators on R3

h i
1 1 −1
1. b. By inspection, A = 2 −1 1 ; by the formulas preceding Theorem 4.4.2, this is the matrix
of projection on y = −x.
72 Vector Geometry

h i
−3 4
d. By inspection, A = 15 4 3 ; by the formulas precedinging Theorem 4.4.2, this is the matrix
of reflection in y = 2x.
h √ i
√1 − 3
f. By inspection, A = 2 1
3 1
; by Theorem 2.6.4 this is the matrix of rotation through π3 .
h i
1 1 m
2. b. For any slope m, projection on the line y = mx has matrix (see the discussion 1+m2 m m2
preceding
h h 4.4.2).iHence the projections on the lines y = x and y = −x have matrices
i Theorem
1 1 1
2 1 1 and 2 −11 −11 , respectively, so the first followed by the second has matrix (note
1

the order) h i h i h i
1 1 −1 1 1 1 1 0 0
2 −1 1 2 1 1 = 4 0 0 =0

It follows that projection on y = x followed by projection on y = −x is the zero transformation.

Note that this conclusion can also be reached geometrically. Given any vector v, its projection
p on the line y = x points along that line. But the line y = −x is perpendicular to the line y = x,
so the projection of p along y = −x will be the zero vector. Since v was arbitrary, this shows
again that projection on y = x followed by projection on y = −x is the zero transformation.
    
17 2 −8 0 26
1 1
3. b. By Theorem 4.4.3: 21 2 20 4 1 = 21 8
−8 4 5 −3 −11
    
22 −4 20 0 −32
1 1
d. By Theorem 4.4.3: 30 −4 28 10 1 = 15 −1
20 10 −20 −3 35
    
9 0 12 1 93
1 1
f. By Theorem 4.4.2: 25 0 0 0 −1 = 25 0
12 0 16 7 124
    
−9 2 −6 2 −28
1 1
h. By Theorem 4.4.2: 11 2 −9 −6 −5 = 11 49
−6 −6 7 0 18
" √ #
3 1
√ 2 −
√2
0
4. b. This is Example 4.4.1 with θ = π6 . Since cos π6 = and sin π6 = 12 , the matrix is
2
3 1
2 2
3
0 =
0 0 1
" √ #  √ √
3 −1 0
     
√ 1 3 √−1 0 1 3
1 3 0 . Hence the rotation of v =
1 1
2
1 0 is 2 1 3 0 0 =2 1 .
0 0 2 3 0 0 2 3 6

6. Denote the rotation by RL, θ . Here the rotation takes place about the y-axis, so RL, θ (j)h= j. In the xz-i
θ − sin θ
plane the effect of RL, θ is to rotate counterclockwise through θ , and this has matrix cos sin θ
h i h i h i h i cos θ
θ sin θ
Theorem 2.6.4. So, in the xz-plane, RL, θ 10 = cos sin θ and RL, θ 01 = −cos θ . Hence
   
cos θ − sin θ
RL, θ (i) = 0 and RL, θ (k) = 0 . Finally, the matrix of
sin θ cos θ
 
  cos θ 0 − sin θ
RL, θ is RL, θ (i) RL, θ (j) RL, θ (k) = 0 1 0 .
sin θ 0 cos θ

 T
9. a. Write v = x y .    h i h i h ih i
v·d ax+by a 1 a2 x + aby 1 a2 ab x
Then PL (v) = proj d v = d= = a2 +b abx + b2 y = a2 +b b2 .
kdk2 a2 +b2 b 2 2 ab y
 4.5. An Application to Computer Graphics 73

h 2 i
1 a ab
Hence the matrix of PL is a2 +b 2 ab b2 . Note that if the line L has slope m this retrieves
h i
1 1 m
the formula 1+m 2 m m2 preceding Theorem 4.4.2. However the present matrix works for
h i
vertical lines, where d = 10 .

4.5 An Application to Computer Graphics

1. b. Translate to the origin, rotate and then translate back. As in Example 4.5.1, we compute
  " √2 − √2 0 # " 1 0 −1 # " 0 6 5 1 3 #
1 0 1 √2 √2
0 1 2 2 2 0 1 −2 0 0 3 3 9
2 2 0
0 0 1 0 0 1 0 0 1 1 1 1 1 1
 √ √ √ √ √ 
√2+2 7√2 + 2 3√2 + 2 −√ 2 + 2 −5√ 2 + 2
1
= 2
−3 2 + 4 3 2+4 5 2+4 2+4 9 2+4
2 2 2 2 2
i h
0
5. b. The line has a point w = , so we translate by −w, then reflect in y = 2x, and then trans-
1
h i
late back by w. The line y = 2x has matrix 51 −34 43 . Thus the matrix (for homogeneous
coordinates) is
      
1 0 0 −3 4 0 1 0 0 −3 4 −4
1 1
0 1 1
5
4 3 0 0 1 −1 =5 4 3 2
0 0 1 0 0 5 0 0 1 0 0 5
      
1 −3 4 −4 1 9
1 1 9 18


Hence for w = 4 we get 5
4 3 2 4 = 5
18 . Hence the point is P 5, 5 .
1 0 0 5 1 5

Supplementary Exercises: Chapter 4

4. Let p and w be the velocities of the airplane and the wind. Then kpk = 100 knots and kwk = 75
knots and the resulting actual velocity of the airplane is v = w + p. Since w and p are orthogonal.
Pythagoras’ theorem gives kvk2 = kwk2 + kpk2 = 752 + 1002 = 252 (32 + 42 ) = 252 · 52 . Hence
kvk = 25 · 5 = 125 knots. The angle θ satisfies cos θ = kwk 75 ◦
kvk = 125 = 0.6 so θ = 0.93 radians or 53 .
 T
6. Let v = x y denote the velocity of the boat in the water. If c is the current velocity then
c = (0, −5) because it flows south at 5 knots. We want to h choose
i v so that the resulting actual
z
velocity w of the boat has easterly direction. Thus w = 0 for some z. Now w = v + c so
h i h i h
z x 0
i h
x
i p √
= + = . Hence z = x and y = 5. Finally, 13 = kvk = x 2 + y2 = x2 + 25
0 y −5 y−5
2
 T
gives x = 144, x = ±12. But x > 0 as w heads east, so x = 12. Thus he steers v = 12 5 , and
the resulting actual speed is kwk = z = 12 knots.
5. The Vector Space Rn

5.1 Subspaces and Spanning

  
0 0
1. b. Yes. In fact, U = span , 0
1 so Theorem 5.1.1 applies.
0 1
     
2 2 4
d. No. 0 is in U but 2 0 = 0 is not in U .
0 0 0
     
0 0 0
f. No. −1 is in U but (−1) −1 = 1 is not in U .
0 0 0

2. b. No. If x = ay + bz equating first and third components gives 1 = 2a + b, 15 = −3b; whence

a = 3, b = −5. This does not satisfy the second component which requires that 2 = −a − b.
d. Yes. x = 3y + 4z.
 
3. b. No. Write these vectors as a1 , a2 , a3 and a4 , and let A = a1 a2 a3 a4 be the matrix
with these vectors as columns. Then det A = 0, so A is not invertible. By Theorem 2.4.5, this
means that the system Ax = b has no solution for some column b. But this says that b is not a
linear combination of the ai by Definition 2.5. That is, the ai do not span R4 .
" 1 #
0
For a more direct proof, 0 is not a linear combination of a1 , a2 , a3 and a4 .
0

10. Since ai xi is in span {xi } for each i, Theorem 5.1.1 shows that span {ai xi } ⊆ span {xi }. Since
xi = a−1
i (ai xi ) is in span {ai xi }, we get span {xi } ⊆ span {ai xi }, again by Theorem 5.1.1.

12. We have U = span {x1 , . . . , xk } so, if y is in U , write y = t1 x1 + · · · + tk xk where the ti are in R.

Then Ay = t1 Ax1 + · · · + t1 Axk = t10 + · · · + tk 0 = 0.

15. b. x = (x + y) − y is in U because x + y and −y = (−1)y are both in U and U is a subspace.

16. b. True. If we take r = 1 we see that x = 1x is in U .

d. True. We have span {y, z} ⊆ span {x, y, z} by Theorem 5.1.1 because both y and z are in
span {x, y, z}. In other words, U ⊆ span {x, y, z}. For the other inclusion, it is clear that y
and z are both in U = span {y, z}, and we are given that x is in U . Hence span {x, y, z} ⊆ U
by Theorem 5.1.1.
nh i h io
1 2
f. False. Every vector in span 0 , 0 has second component zero.

75
76 The Vector Space Rn

20. If U is a subspace then S2 and S3 certainly hold. Conversely, suppose that S2 and S3 hold. It is here
that we need the condition that U is nonempty. Because we can then choose some x in U , and so
0 = 0x is in U by S3. So U is a subspace.

22. b. First, 0 is in U +W because 0 = 0 + 0 (and 0 is in both U and W ). Now suppose that P and Q
are both in U +W , say p = x1 + y1 and q = x2 + y2 where x1 and x2 are in U , and y1 and y2
are in W . Hence

p + q = (x1 + y1 ) + (x2 + y2 ) = (x1 + x2 ) + (y1 + y2 )

so p + q is in U + W because x1 + x2 is in U (both x1 and x2 are in U ), and y1 + y2 is in W .

Similarly
aP = a(x1 + y1 ) = ax1 + ay1
is in p + q because ax1 is in p and ay1 is in Q. Hence U +W is a subspace.

5.2 Independence and Dimension

1. b. Yes. The matrix with these vectors as columns has determinant −2 6= 0, so Theorem 5.2.3
applies.
d. No. (1, 1, 0, 0) − (1, 0, 1, 0) + (0, 0, 1, 1) − (0, 1, 0, 1) = (0, 0, 0, 0) is a nontrivial linear
combination that vanishes.

2. b. Yes. If a(x + y) + b(y + z) + c(z + x) = 0 then (a + c)x + (a + b)y + (b + c)z = 0. Since we are
assuming that {x, y, z} is independent, this means a + c = 0, a + b = 0, b + c = 0. The only
solution is a = b = c = 0.
d. No. (x + y) − (y + z) + (z + w) − (w + x) = 0 is a nontrivial linear combination that vanishes.

3. b. Write x1 = (2, 1, 0, −1), x2 = (−1, 1, 1, 1), x3 = (2, 7, 4, 1), and write U = span {x1 , x2 , x3 }.
Observe that x3 = 3x1 + 4x2 so U = {x1 , x2 }. This is a basis because {x1 , x2 } is independent,
so the dimension is 2.
d. Write x1 = (−2, 0, 3, 1), x2 = (1, 2, −1, 0), x3 = (−2, 8, 5, 3), x4 = (−1, 2, 2, 1) and write
U = span {x1 , x2 , x3 , x4 }. Then x3 = 3x1 +4x2 and x4 = x1 +x2 so the space is span {x1 , x2 }.
As this is independent, it is a basis so the dimension is 2.

4. b. (a +b, a −b, b, a) = a(1, 1, 0, 1) +b(1, −1, 1, 0) so U = span {(1, 1, 0, 1), (1, −1, 1, 0)}.
This is a basis so dim U = 2.
d. (a − b, b + c, a, b + c) = a(1, 0, 1, 0) + b(−1, 1, 0, 1) + c(0, 1, 0, 1). Hence U =
span {(1, 0, 1, 0), (−1, 1, 0, 1), (0, 1, 0, 1)}. This is a basis so dim U = 3.
f. If a + b = c + d then a = −b + c + d. Hence U = {(−b + c + d, b, c, d) | b, c, d in R} so
U = span {(−1, 1, 0, 0), (1, 0, 1, 0), (1, 0, 0, 1)}. This is a basis so dim U = 3.
 5.2. Independence and Dimension 77

5. b. Let a(x + w) + b(y + w) + c(z + w) + dw = 0, that is ax + by + cz + (a + b + c + d)w = 0. As

{x, y, z, w} is independent, this implies that a = 0, b = 0, c = 0 and a + b + c + d = 0. Hence
d = 0 too, proving that {x + w, y + w, z + w, w} is independent. It is a basis by Theorem 5.2.7
because dim R4 = 4.
6. b. Yes. They are independent (the matrix with them as columns has determinant −2) and so are
a basis of R3 by Theorem 5.2.7 (since dim R3 = 3).
d. Yes. They are independent (the matrix with them as columns has determinant −6) and so are
a basis of R3 by Theorem 5.2.7 (since dim R3 = 3).
f. No. The determinant of the matrix with these vectors as its columns is zero, so they are not
independent (by Theorem 5.2.3). Hence they are not a basis of R4 because dim R4 = 4.
7. b. True. If sy + tz = 0 then 0x + sy + tz = 0, so s = t = 0 by the independence of {x, y, z}.
d. False. If x 6= 0 let k = 2, x1 = x and x2 = −x. Then each xi 6= 0 but {x1 , x2 } is not independent.
f. False. If y = −x and z = 0 then 1x + 1y + 1z = 0, but {x, y, z} is certainly not independent.
h. True. The xi are not independent so, by definition, some nontrivial linear combination vanishes.
10. If rx2 + sx3 +tx5 = 0 then 0x1 + rx2 + sx3 + 0x4 +tx5 + 0x6 = 0. Since the larger set is independent,
this implies r = s = t = 0.
12. If t1 x1 + t2 (x1 + x2 ) + · · · + tk (x1 + x2 + · · · + xk ) = 0 then, collecting terms in x1 , x2 , . . . ,
(t1 + t2 + · · · + tk )x1 + (t2 + · · · + tk )x2 + · · · + (tk−1 + tk )xk−1 + tk xk = 0
Since {x1 , x2 , . . . , xk } is independent we get
t1 + t2 + · · · + tk = 0
t2 + · · · + tk = 0
..
.
tk−1 + tk = 0
tk = 0
The solution (from the bottom up) is tk = 0, tk−1 = 0, . . . , t2 = 0, t1 = 0.
b. We show that hAT is T 2
16. i invertible. Suppose A x = 0x in R . By Theorem 2.4.5, we must show that
x = 0. If x = st then AT x = 0 gives as + ct = 0, bs + dt = 0. But then s(ax + by) + t(cx +
dy) = (sa +tc)x + (sb +td)y = 0. Hence s = t = 0 because {ax + by, cx + dy} is independent.
17. b. Note first that each V −1 xi is in null (AV ) because (AV )(V −1 xi ) = Axi = 0. If t1V −1 x1 + · · · +
tkV −1 xk = 0 then V −1 (t1x1 + · · · + tk xk ) = 0 so t1 x1 + · · · + tk xk = 0 
(by multiplication by
V ). Thus t1 = · · · = tk = 0 because {x1 , . . . , xk } is independent. So V −1 x1 , . . . , V −1 xk
is independent. To see that it spans null (AV ), let y be in null (AV ), so that AV y = 0. Then
V y is in null A so V y = s1 x1 + · · · + sn xn because {x1 , . . . , xn } spans null A. Hence y =
s1V −1 x1 + +skV −1 xk , as required.
20. We have {0} ⊆ U ⊆ W where dim {0} = 0 and dim W = 1. Hence dim U is an integer between 0
and 1 (by Theorem 5.2.8), so dim U = 0 or dim U = 1. If dim U = 0 then U = {0} by Theorem
5.2.8 (because {0} ⊆ U and both spaces have dimension 0); if dim U = 1 then U = W again by
Theorem 5.2.8 (because U ⊆ W and both spaces have dimension 1).
78 The Vector Space Rn

5.3 Orthogonality
n o
1. b. √1 (1,
1, 1), √1 (4, 1, −5), √1 (3, −3, 1) where in each case we divide by the norm of
3 42 14
the vector.

3. b. Write e1 = (1, 0, −1), e2 = (1, 4, 1), e3 = (2, −1, 2). Then

e1 · e2 = 1 + 0 − 1 = 0, e1 · e3 = 2 + 0 − 2 = 0, e2 · e3 = 2 − 4 + 2 = 0

so {e1 , e2 , e3 } is orthogonal and hence a basis of R3 . If x = (a, b, c), Theorem 5.3.6 gives
x·e1
x= e1 + x·e22 e2 + x·e32 e3 = a−c
2 e1 +
a+4b+c
18 e2 +
2a−b+2c
e3
ke1 k2 ke2 k ke3 k 9

d. Write e1 = (1, 1, 1), e2 = (1, −1, 0), e3 = (1, 1, −2). Then

e1 · e2 = 1 − 1 + 0 = 0, e1 · e3 = 1 + 1 − 2 = 0, and e2 · e3 = 1 − 1 + 0 = 0

Hence {e1 , e2 , e3 } is orthogonal and hence is a basis of R3 . If x = (a, b, c), Theorem 5.3.6
gives
x = x·e12 e1 + x·e22 e2 + x·e32 e3 = a+b+c a−b
3 e1 + 2 e2 +
a+b−2c
6 e3
ke1 k ke2 k ke3 k

4. b. If e1 = (2, −1, 0, 3) and e2 = (2, 1, −2, −1) then {e1 , e2 } is orthogonal because e1 · e2 =
4 − 1 + 0 − 3 = 0. Hence {e1 , e2 } is an orthogonal basis of the space U it spans. If x =
(14, 1, −8, 5) is in U , Theorem 5.3.6 gives
x·e1
x= e1 + x·e22 e2 = 42 40
14 e1 + 10 e2 = 3e1 + 4e2
ke1 k2 ke2 k

We check that these are indeed equal. [We shall see in Section 8.1 that in any case,
 
x·e1 x·e2
x− 2 e1 + 2 e2 is orthogonal to every vector in U .]
ke1 k ke2 k

5. b. The condition that (a, b, c, d) is orthogonal to each of the other three vectors gives the fol-
lowing equations for a, b, c, and d.

a − c + d = 0
2a + b + c − d = 0
a − 3b + c = 0

Solving we get:    
1 0 −1 1 0 1 0 −1 1 0
2 1 1 −1 0 → 0 1 3 −3 0
1 −3 1 0 0 1 −3 2 −1 0
" 1
#
1 0 0 0
 
1 0 −1 1 0 11
→ 0 1 3 −3 0 → 0 1 0 3
− 11 0
0 0 11 −10 0 0 0 1 10
− 11 0

The solution is (a, b, c, d) = t(−1, 3, 10, 11), t in R.

 5.3. Orthogonality 79

6. b. k2x + 7yk2 = (2x + 7y) · (2x + 7y)

= 4(x · x) + 14(x · y) + 14(y · x) + 49(y · y)
= 4 kxk2 + 28(x · y) + 49 kyk2
= 36 − 56 + 49
= 29
d. (x − 2y) · (3x + 5y) = 3(x · x) + 5(x · y) − 6(y · x) − 10(y · y)
= 3 kxk2 − (x · y) − 10 kyk2
= 27 + 2 − 10
= 19

7. b. False. For example, if x = (1, 0) and y = (0, 1) in R2 , then {x, y} is orthogonal but x + y =
(1, 1) is not orthogonal to x.
d. True. Let x and y be distinct vectors in the larger set. Then either both are xi ’s, both are yi ’s, or
one is an xi and one is a yi . In the first two cases x · y = 0 because {xi } and {y j } are orthogonal
sets; in the last case x · y = 0 by the given condition.
f. True. Every pair of distinct vectors in {x} are orthogonal (there are no such pairs). As x 6= 0,
this shows that {x} is an orthogonal set.

9. Row i of AT is cTi so the (i, j) entry of AT A is cTi c j = ci · c j . This is 0 if i 6= j, and 1 if i = j. That is

AT A = I.

11. b. Take x = (1, 1, 1)qand y = (r1 , r2 , r3 ). Then |x · y| ≤ kxk kyk by Theorem 5.3.2; that is
√
|r1 + r2 + r3 | ≤ 3 r12 + r22 + r32 . Squaring both sides gives

r12 + r22 + r32 + 2(r1 r2 + r1 r3 + r2 r3 ) ≤ 3(r12 + r22 + r32 )

Simplifying we obtain r1 r2 + r1 r3 + r2 r3 ≤ r12 + r22 + r32 , as required.

12. b. Observe first that

(x + y) · (x − y) = kxk2 − kyk2 (∗)
holds for all vectors x and y in Rn .
If x + y and x − y are orthogonal then (x + y) · (x − y) = 0, so kxk2 = kyk2 by (∗). Taking
positive square roots gives kxk = kyk.
Conversely, if kxk = kyk then certainly kxk2 = kyk2 , so (∗) gives (x + y) · (x − y) = 0. This
means that x + y and x − y are orthogonal.

15. If λ is an eigenvalue of AT A, let (AT A)x = λ x where x 6= 0 in Rn . Then:

kAxk2 = (Ax)T (Ax) = (xT AT )Ax = xT (AT Ax) = xT (λ x) = λ kxk2

kAxk2
6 0 (because x 6= 0), this gives λ =
Since kxk = kxk2
≥ 0.
80 The Vector Space Rn

5.4 Rank of a Matrix

− 21 1
 
" 2 −1 1
# " 2 −1 1
# 1 2
−2 1 1 0 0 2 0 0 1
1. b. 4 −2 3 → 0 0 1 → 0 0 0
.
−6 3 0 0 0 3 0 0 0
n T  T o
Hence, rank A = 2 and 1 − 12 1
2 , 0 0 1 is a basis of row A. Thus
n T  T o
2 −1 1 , 0 0 1 is also a basis of row A. Since the leading 1’s are in columns
1 and 3, columns 1 and 3 of A are a basis of col A.
h i h i h i
d. −31 −62 −13 −23 → 01 02 −10 73 → 10 20 −10 31 .
n T  T o
Hence, rank A = 2 and 1 2 −1 3 , 0 0 0 1 is a basis of row A. Since the
leading 1’s are in columns 1 and 4, columns 1 and 4 of A are a basis of col A.

2. b. Apply the gaussian algorithm to the matrix with these vectors as rows:
" 1 −1
# " 2 5
#  1 1 1 0 0 0 1 1 0 0 0

3 1 4 2 7 0 −2 2 5 1 0 1 −1 − 25 − 21
1 1 0 0 0 → 0 −2 4 2 7 → 0 0 1 − 32 3

5 1 6 7 8 0 −4 6 7 8 0 0 0 0 0

n T  T  T o
Hence, 1 1 0 0 0 , 0 2 −2 −5 −1 , 0 0 2 −3 6 is a basis
of U (where we have cleared fractions using scalar multiples).
d. Write these columns as the rows of the following matrix:
" 1 5 −6 # " 1 5 −6 # " 1 5 −6
# " 1 5 −6
#
2 6 −8 0 −4 4 0 1 −1 0 1 −1
3 7 −10 → 0 −8 8 → 0 0 24 → 0 0 1
4 8 12 0 −12 36 0 0 0 0 0 0

     
1 0 0
Hence, 5 , 1 , 0 is a basis of U .
−6 −1 1

3. b. No. If the 3 columns were independent, the rank would be 3.

No. If the 4 rows were independent, the rank would be 4, a contradiction here as the rank
cannot exceed the number of columns.
d. No. Suppose that A is m × n. If the rows are independent then rank A = dim ( row A) = m (the
number of rows). Similarly if the columns are independent then rank A = n (the number of
columns). So if both the rows and columns are independent then m = rank A = n, that is A is
square.
f. No. Then dim ( null A) = n − r = 4 − 2 = 2, contrary to null (A) = Rx where x 6= 0.
T
∈ Rn then Ax = x1 c1 + · · · + xn cn by Definition

4. Let c j denote column j of A. If x = x1 · · · xn
2.5. Hence

col A = span {c1 , . . . , cn } = {x1 c1 + · · · + xn cn | x j ∈ R} = {Ax | x ∈ Rn }

 5.4. Rank of a Matrix 81

7. b. The null space of A is the set of columns x such that Ax = 0. Applying gaussian elimination to
the augmented matrix gives:
" 3 5 5 2 0 0 # " 1 0 2 2 1 0 #
1 0 2 2 1 0 0 5 −1 −4 −3 0
1 1 1 −2 −2 0 → 0 1 −1 −4 −3 0
−2 0 −4 −4 −2 0 0 0 0 0 0 0

1 0 2 2 1 0 1 0 0 −6 −5 0
" # " #
0 1 −1 −4 −3 0 0 1 0 0 0 0
→ 0 0 4 16 12 0 → 0 0 1 4 3 0
0 0 0 0 0 0 0 0 0 0 0 0
 6s + 5t  
 0 
Hence, the set of solutions is null A =  −4s − 3t  | s, t in R = span B where
 s 
t
 6   5 
 0 0 
B =  −4  ,  −3  . Since B is independent, it is the required basis of null A. We have
 1 0 
0 1
r = rank A = 3 by the above reduction, so n − r = 5 − 3 = 2. This is the dimension of null A,
as Theorem 5.4.3 asserts.
 
8.  A. To compute rank A, let R = r1 r2 · · · rn .
b. Since A is m × n, dim ( null A) = n − rank
Then A = CR = r1C r2C · · · rnC by block multiplication, so
col A = span {r1C, r2C, . . . , rnC} = span {C}
because some ri 6= 0. Hence rank A = 1, so dim ( null A) = n − rank A = n − 1.
b. Let A = c1 · · · cn where c j is the jth column of A; we must show that {c1 , . . . , cn } is
 
9.
 T
independent. Suppose that x1 c1 + · · · + xn cn = 0, xi in R. If we write x = x1 · · · xn ,
this reads Ax = 0 by Definition 2.5. But then x is in null A, and null A = 0 by hypothesis. So
x = 0, that is each xi = 0. This shows that {c1 , . . . , cn } is independent.
10. b. If A2 = 0 then A(Ax) = 0 for all x in Rn , that is {Ax | x in Rn } ⊆ null A. But col A = {Ax | x
in Rn }, so this shows that col A ⊆ null A. If we write r = rank A, taking dimensions gives
r = dim ( col A) ≤ dim ( null A) = n − r by Theorem 5.4.3. It follows that 2r ≤ n; that is r ≤ n2 .
12. We have rank (A) = dim [ col (A)] and rank (AT ) = dim [ row (AT )]. Let {c1 , c2 , . . . , ck } be a basis
of col(A); it suffices to show that {cT1 , cT2 , . . . , cTk } is a basis of row (AT ). But if t1 cT1 + t2 cT2 +
· · · + tk cTk = 0, t j in R, then (taking transposes) t1 c1 + t2 c2 + · · · + tk ck = 0 so each t j = 0. Hence
{cT1 , cT2 , . . . , cTk } is independent. Given v in row (AT ) then vT is in col (A), say vT = s1 c1 + s2 c2 +
· · · + sk ck , s j in R. Hence v = s1 cT1 + s2 cT2 + · · · + sk cTk so {cT1 , cT2 , . . . , cTk } spans row (AT ), as
required.
15. b. Let {c1 , . . . , cr } be a basis of col A where r = rank A. Since Ax = b has no solution, b is not
in col A = span {c1 , · · · , cr } by Exercise 12. It follows that {c1 , . . . , cr , b} is independent
[If a1 c1 + · · · + ar cr + ab = 0 then a = 0 (since b is not in col A), whence each ai = 0 by the
independence of the ci ]. Hence, it suffices to show that col [A, B] = span {c1 , · · · , cr , b}. It
is clear that b is in col [A, b], and each c j is in col [A, b] because it is a linear combination of
columns of A (and so those of [A, b]). Hence
span {c1 , . . . , cr , b} ⊆ col [A, b]
82 The Vector Space Rn

On the other hand, each column x in col [A, b] is a linear combination of b and the columns
of A. Since these columns are themselves linear combinations of the c j , so x is a linear combi-
nation of b and the c j . That is, x is in span {c1 , . . . , cr , b}.

5.5 Similarity and Diagonalization

1. b. det A = −5, det B = −1 (so A and B are not similar). However, tr A = 2 = tr B, and rank A =
2 = rank B (both are invertible).
d. tr A = 5, tr B = 4 (so A and B are not similar). However, det A = 7 = det B, so rank A = 2 =
rank B (both are invertible).
f. tr A = −5 = tr B; det A = 0 = det B; however rank A = 2, rank B = 1 (so A and B are not
similar).

3. b. We have A ∼ B, say B = P−1 AP. Hence B−1 = (P−1 AP)−1 = P−1 A−1 (P−1 )−1 , so A−1 ∼ B−1
because P−1 is invertible.
x−3 0 −6
4. b. cA (x) = 0 x+3 0 = (x + 3)(x2 − 5x − 24) = (x + 3)2 (x − 8). So the eigenvalues are
−5 0 x−2
λ1 = −3, λ2 = 8. To find  the associated eigenvectors:    
−6 0 −6 1 0 1 −1 0
λ1 = −3: 0 0 0 → 0 0 0 ; basic eigenvectors 0 , 1 .
 −5 0 −5  0 0 06   1 0
5 0 −6 1 0 −5 6
λ2 = 8: 0 11 0 → 0 1 0 ; basic eigenvector 0 .
 −5  0  6    0 0 0 5
−1 0 6
Since 0 , 1 , 0 is a basis of eigenvectors, A is diagonalizable and
1
  0 5
 
−1 0 6 −3 0 0
P= 0 1 0 −1
will satisfy P AP = 0 −3 0 .
1 0 5 0 0 8
   
x−4 0 0 0 0 0 1 0 0
2
d. cA (x) = 0 x − 2 −2 = (x − 4) (x + 1). For λ = 4, 0 2 −2 → 0 1 −1 ;
−3 x−1 2 −3
 2 3 0 0 0
0
E1 = 1 . Hence A is not diagonalizable by Theorem 5.5.6 because the dimension of E4 (A) =
1
1 while the eigenvalue 4 has multiplicity 2.

8. b. If B = P−1 AP and Ak = 0, then Bk = (P−1 AP)k = P−1 Ak P = P−1 0P = 0.

9. b. Let the diagonal entries of A all equal λ . If A is diagonalizable then P−1 AP = λ I by Theorem
5.5.3 for some invertible matrix P. Hence A = P(λ I)P−1 = λ (PIP−1) = λ I.

10. b. Let P−1 AP = D = diag {λ1 , λ2 , . . . , λn }. Since A and D are similar matrices, they have the
same trace by Theorem 5.5.1. That is

tr A = tr (P−1 AP) = tr D = λ1 + λ2 + · · · + λn

12. b. TP (A)TP (B) = (P−1 AP)(P−1 BP) = P−1 AIBP = P−1 ABP = TP (AB)
 5.6. Best Approximation and Least Squares 83

13. b. Assume that A is diagonalizable, say A ∼ D where D is diagonal, say D = diag (λ1 , . . . , λn )
where λ1 , · · · , λn are the eigenvalues of A. But A and AT have the same eigenvalues (Example
3.3.5) so AT ∼ D also. Hence A ∼ D ∼ AT , so A ∼ AT as required.

17. b. We use Theorem 5.5.7. The characteristic polynomial of B is computed by first adding rows 2
and 3 to row 1. For convenience, write s = a + b + c, k = a2 + b2 + c2 − (ab + ac + bc).
x−c −a −b x−s x−s x−s x−s 0 0
cB (x) = −a x−b −c = −a x−b −c = −a x + (a − b) a−c
−b −c x−a −b −c x−a −b b−c x − (a − b)

= (x − s) x2 − (a − b)2 − (a − c)(b − c)
 

= (x − s)(x2 − k)
√ √
Hence, the eigenvalues of B are s, k and − k. These must be real by Theorem 5.5.7, so
k ≥ 0. Thus a2 + b2 + c2 ≥ ab + ac + bc.

20. b. To compute cA (x) = det (xI − A), add x times column 2 to column 1, and expand along row 1:
x −1 0 0 ··· 0 0
0 x −1 0 ··· 0 0
0 0 x −1 ··· 0 0
.. .. .. .. .. ..
cA (x) = . . . . . .
0 0 0 0 ··· −1 0
0 0 0 0 ··· x −1
−r0 −r1 −r2 −r3 ··· −rk−2 x − rk−1
0 −1 0 0 ··· 0 0
x2 x −1 0 ··· 0 0
0 0 x −1 ··· 0 0
. . . . . .
= .. .. .. .. .. ..
0 0 0 0 ··· −1 0
0 0 0 0 x −1
−r0 − r1 x −r1 −r2 −r3 −rk−2 x − rk−1

Now expand along row 1 to get

x2 −1 ··· 0 0
0 x −1 ··· 0 0
. . . .
.. .. .. ..
cA (x) =
0 0 0 ··· −1 0
0 0 0 ··· x −1
−r0 − r1 x −r2 −r3 ··· −rk−2 x − rk−1

This matrix has the same form as xI − A, so repeat this procedure. It leads to the given expres-
sion for det (xI − A).

5.6 Best Approximation and Least Squares

3 1 1 6
" # " #    
x 26 −2 12
2 3 −1 1
1. b. Here A = 2 −1 1 ,b= 0 ,x= y . Hence, AT A = −2 20 −12 .
z 12 −12 12
3 −3 3 8
84 The Vector Space Rn

This is invertible and the inverse is

   
96 −120 −216 24 −30 −54
T −1 1 1
(A A) = 144 −120 168 288 = 36
−30 42 72
−216 288 516 −54 72 129

Here the (unique) best approximation is

      
24 −30 −54 44 −60 −20
T −1 T 1 1 1
z = (A A) A b = 36 −30 42 72 −15 = 36
138 = 12
46
−54 72 129 29 285 95

Of course this can be found more efficiently using gaussian elimination on the normal equa-
tions for z.
i 1 2 i 4
" # " #
h h i h h i
2. b. Here M T M = 12 14 17 18 1 4
1 7 = 4 21
21 133 , M Ty = 1 1 1 1
2 4 7 8
3
2 = 10
42 . We
1 8 1
solve the normal equation (M T M)A = M T y by inverting M T M:
h ih i h i h i
133 −21
A = (M T M)−1 M T y = 91
1
−21 4
10
42 = 1
91 · 448
−42 = 1
13
64
−6

64 6
Hence the best fitting line has equation y = 13 − 13 x.
4
d. Analogous to (b). The best fitting line is y = − 10 − 17
10 x.

3. b. Now
" 1 −2 4
 #  
1 1 1 1 4 5 29
T 1 0 0
M M= −2 0 3 4
1 3 9 = 5 29 83
4 0 9 16 29 83 353
1 4 16
" 1
 #  
1 1 1 1 6
T 0
M y= −2 0 3 4
2 = 16
4 0 9 16 70
3

We use (MM T )−1 to solve the normal equations even though it is more efficient to solve them
by gaussian elimination.
      
3348 642 −426 6 540 .127
T −1 T 1 1
A = (M M) (M y) = 4248 642 571 −187 16 = 4248 −102 = −.024
−426 −187 91 70 822 .194

Hence the best fitting quadratic has equation y = .127 − .024x + .194x2 .
" 1 # " 0 02 20 # " 0 0 1
#
1 1 12 21 1 1 2
4. b. In the notation of Theorem 5.6.3: y = ,M= 5 2 22 22 = 2 4 4 .
10 3 32 23 3 9 8
     
14 36 34 230 0 −92 115 0 −46
T T −1 1 1
Hence, M M = 36 98 90 , and (M M) = 92 0 34 −36 = 46
0 17 −18 .
34 90 85 −92 −36 76 −46 −18 38
Thus, the (unique) solution to the normal equation is
    
115 0 −46 41 −23
T −1 T 1 1
z = (M M) M y = 46 0 17 −18 111 = 46
33
−46 −18 38 103 30

1 2 x
The best fitting function is thus 46 [−23x + 33x + 30(2) ].
 5.6. Best Approximation and Least Squares 85

1
(−1)2 sin − π2
  
  
2 1 1 1 −1
1 1 02 sin(0) 1 0 0
5. b. Here y =  5
, M = 
1 22 sin(π )
=
1 4 0
. Hence
32 sin 32π


9 1 1 9 −1
   
4 14 0 −24 7 35
T T −1 1
M M= 14 98 −10 and (M M) = 2
7 −2 −10
0 −10 2 35 −10 −49

Thus, the (unique) solution to the normal equations is

 " 31
# "
18
#
24 −2 14 2
z = (M T M)−1 M T y = 1
40
−2 12 3 203
2 = 1
20
21
14 3 49 − 19 28
2

Hence, the best fitting functions

1 2 πx


20 [18 + 21x + 28 sin 2 ]

7. To fit s = a + bx where x = t 2, we have

h i 1 1
 h i
1 1 1 3 14
MT M = 1 4 9
1 4 = 14 98
1 9
h i 95
 h i
T 1 1 1 231
M y= 1 4 9
80 = 919
56
h ih i h i h i
1 98 −14 231 1 9772 99.71
Hence A = (M T M)−1 M T y = 98 −14 3 919 = 98 −477 = −4.87 to two decimal places.
Hence the best fitting equation is

y = 99.71 − 4.87x = 99.71 − 4.87t 2

Hence the estimate for g comes from − 12 g = −4.87, g = 9.74 (the true value of g is 9.81).
Now fit s = a + bt + ct 2 . In this case
    
1 1 1 1 1 1 3 6 14
T
M M= 1 2 3 1 2 4 = 6 14 36
1 4 9 1 3 9 14 36 98
    
1 1 1 95 231
MT y = 1 2 3 80 = 423
1 4 9 56 919

Hence " #
101
    
76 −84 20 231 404
A = (M T M)−1 (M T y) = 1
4
−84 98 −24 423 = 1
4
−6 = − 32
20 −24 6 919 −18 − 92

so the best quadratic is y = 101 − 23 t − 92 t 2 . This gives − 92 = − 12 g so the estimate for g is g = 9 in

this case.

9. We want r0 , r1 , r2 , and r3 to satisfy

r0 + 50r1 + 18r2 + 10r3 = 28

r0 + 40r1 + 20r2 + 16r3 = 30
r0 + 35r1 + 14r2 + 10r3 = 21
86 The Vector Space Rn

r0 + 40r1 + 12r2 + 12r3 = 23

r0 + 30r1 + 16r2 + 14r3 = 23

We settle for a best approximation. Here

 1 50 18 10
  28

1 40 20 16 30
A= 1 35 14 10  b= 21 
1 40 12 12 23
1 30 16 14 23

5 195 80 62
" #
195 7825 3150 2390
AT A = 80 3150 1320 1008
62 2390 1008 796
1035720 −16032 10080 −45300
" #
−16032 416 −632 800
(AT A)−1 = 1
50160 10080 −632 2600 −2180
−45300 800 −2180 3950

So the best approximation

1035720 −16032 10080 −45300 125 −5.19
" #" # " #
−16032 −632
z = (AT A)−1 (AT b) = 1
50160 10080
416
−632 2600
800
−2180
4925
2042 = 0.34
0.51
−45300 800 −2180 3950 1568 0.71

The best fitting function is

y = −5.19 + 0.34x1 + 0.51x2 + 0.71x3

10. b. f (x) = a0 here so the sum of squares is

s = (y1 − a0 )2 + (y2 − a0 )2 + · · · + (yn − a0 )2

n
= ∑ (yi − a0 )2
i=1
n
= ∑ (a20 − 2a0 yi + y2i )
i=1
na20 − 2 ∑ yi a0 + ∑ y2i



=

— a quadratic in a0 . Completing the square gives

2 h
2 i
s = n a0 − 1n ∑ yi − ∑ y2i − 1n ∑ yi


This is minimal when a0 = 1n ∑ yi .

ex1 ex1
" 1
# " 1
# " #
. . . .
13. b. It suffices to show that the columns of M = .. .. are independent. If r0 .. +r1 .. =
1 x
en 1 x
e2
" 0 #
..
. , then r0 + r1 exi = 0 for each i. Thus, r1 (exi − ex j ) = 0 for all i and j, so r1 = 0 because
0
two xi are distinct. Then r0 = r1 ex1 = 0 too.
 5.7. An Application to Correlation and Variance 87

5.7 An Application to Correlation and Variance

   
2. Let x = x1 x2 · · · x10 = 12 16 13 · · · 14 denote the number of years of education.
1 1
Then x = 10 ∑ xi = 15.3, and s2x = n−1 ∑(xi − x)2 = 9.12 (so sx = 3.02).
   
Let y = y1 y2 · · · y10 = 31 48 35 · · · 35 denote the number of dollars (in thou-
1 1
sands) of yearly income. Then y = 10 ∑ ti = 40.3, and s2y = n−1 ∑(yi − y)2 = 114.23 (so sy = 10.69).
x·y−10xy
The correlation is r = 9sx sy = 0.599.

4. b. We have zi = a + bxi for each i, so

1
∑(a + bxi) = 1n na + b ∑ xi = a + b 1
∑ xi



z= n n = a + bx

Hence

s2z = 1
n−1 ∑(zi − z)2 = n−1
1
∑[(a + bxi) − (a + bx)]2 = n−1
1
∑ b2 (xi − x)2 = b2s2x
√
The result follows because b2 = |b|.

Supplementary Exercises: Chapter 5

1. b. False. If r = 0 then rx is in U for any x.

d. True. If x is in U then −x = (−1)x is also in U by axiom S3 in Section 5.1.
f. True. If rx + sy = 0 then rx + sy + 0z = 0 so r = s = 0 because {x, y, z} is independent.
h i h i
h. False. Take n = 2, x1 = 11 and x2 = −1 −1 . Then both x1 and x2 are nonzero, but {x1 , x2 }
is not independent.
j. False. If a = b = c = 0 then ax + by + cz = 0 for any x, y and z.
l. True. If t1 x1 + t2 x2 + · · · + tn xn = 0 implies that each ti = 0, then {x1 , x2 , . . . , xn } is indepen-
dent, contrary to assumption.
(" 1 # " −1 # " 0 # " 0 #)
0 0 0 0
n. False. 0 , 0 , 1 , 0 is not independent.
0 0 0 1

p. False. {x, x + y, y} is never independent because 1x + (−1)(x + y) + 1y = 0 is a nontrivial

vanishing linear combination.
r. False. Every basis of R3 must contain exactly 3 vectors (by Theorem 5.2.5). Of course a
nonempty subset of a basis will be independent, but it will not span R3 if it contains fewer than
3 vectors.
6. Vector Spaces

6.1 Examples and Basic Properties

1. b. No: S5 fails 1(x, y, z) = (1x, 0, 1z) = (x, 0, z) 6= (x, y, z) for all (x, y, z) in V . Note that the
other nine axioms do hold.
d. No: S4 and S5 fail: S5 fails because 1(x, y, z) = (2x, 2y, 2z) 6= (x, y, z); and S4 fails because
a[b(x, y, z)] = a(2bx, 2by, 2bz) = (4abx, 4aby, 4abz) 6= (2abx, 2aby, 2abz) = ab(x, y, z).
Note that the eight other axioms hold.
2. b. No: A1 fails — for example (x3 + x + 1) + (−x3 + x + 1) = 2x + 2 is not in the set.
d. No: A1 and S1 both fail. For example x + x2 and 2x are not in the set. Hence none of the other
axioms make sense. h i h i
f. Yes. First verify A1 and S1. Suppose A = ac db and B = xz y
w are in V , so a + c = b + d
h i
+x b+y
and x + z = y + w. Then A + B = ac + z d+w is in V because

(a + x) + (c + z) = (a + c) + (x + z) = (b + d) + (y + w) = (b + y) + (d + w)
h i
Also rA = ra rb
rc rd is in V for all r in R because ra + rc = r(a + c) = r(b + d) = rb + rd.
A2, A3,
h S2,iS3, S4, S5. These hold for matrices in general.
A4. 00 00 is in V and so serves as the zero of V .
h i h i
A5. Given A = ac db with a+c = b+d, then −A = −a −b
−c −d is also in V because −a−c =
−(a + c) = −(b + d) = −b − d. So −A is the negative of A in V .
h. Yes. The vector space axioms are the basic laws of arithmetic.
j. No. S4 and S5 fail. For S4, a(b(x, y)) = a(bx, −by) = (abx, aby), and this need not equal
ab(x, y) = (abx, −aby); as to S5, 1(x, y) = (x, −y) 6= (x, y) if y 6= 0.
Note that the other axioms do hold here:
A1, A2, A3, A4 and A5 hold because they hold in R2 .
S1 is clear; S2 and S3 hold because they hold in R2 .
l. No. S3 fails: Given f : R → R and a, b in R, we have

[(a + b) f ](x) = f ((a + b)x) = f (ax + bx)

(a f + b f )(x) = (a f )(x) + (b f )(x) = f (ax) + f (bx)

These need not be equal: for example, if f is the function defined by f (x) = x2 ;
Then f (ax + bx) = (ax + bx)2 need not equal (ax)2 + (bx)2 = f (ax) + f (bx).

89
90 Vector Spaces

Note that the other axioms hold. A1-A4 hold by Example 6.1.7 as we are using pointwise
addition.
S2. a( f + g)(x) = ( f + g)(ax) definition of scalar multiplication in V
= f (ax) + g(ax) definition of pointwise addition
= (a f )(x) + (ag)(x) definition of scalar multiplication in V
= (a f + ag)(x) definition of pointwise addition
As this is true for all x, a( f + g) = a f + ag.
S4. [a(b f )](x) = (b f )(ax) = f [b(ax)] = f [(ba)x] = [(ba) f ](x) = [ab f ](x) for all x,
so a(b f ) = (ab) f .
S5. (1 f )(x) = f (1x) = f (x) for all x, so 1 f = f .
n. No. S4, S5 fail: a ∗ (b ∗ X ) = a ∗ (bX T ) = a(bX T )T = abX T T = abX , while (ab) ∗ X = abX T .
These need not be equal. Similarly: 1 ∗ X = 1X T = X T need not equal X .
Note that the other axioms do hold:
A1-A5. These hold for matrix addition generally.
S1. a ∗ X = aX T is in V .
S2. a ∗ (X +Y ) = a(X +Y )T = a(X T +Y T ) = aX T + aY T = a ∗ X + a ∗Y .
S3 (a + b) ∗ X = (a + b)X T = aX T + bX T = a ∗ X + b ∗ X .

4. A1. (x, y) + (x1 , y1 ) = (x + x1 , y + y1 + 1) is in V for all (x, y) and (x1 , y1 ) in V .

A2. (x, y) + (x1 , y1 ) = (x + x1 , y + y1 + 1) = (x1 + x, y1 + y + 1) = (x1 , y1 ) + (x1 , y).
A3. (x, y) + ((x1 , y1 ) + (x2 , y2 )) = (x, y) + (x1 + x2 , y1 + y2 + 1)
= (x + (x1 + x2 ), y + (y1 + y2 + 1) + 1)
= (x + x1 + x2 , y + y1 + y2 + 2)
((x, y) + (x1 , y1 )) + (x2, y2 ) = (x + x1 , y + y1 + 1) + (x2 , y2 )
= ((x + x1 ) + x2 , (y + y1 + 1) + y2 + 1)
= (x + x1 + x2 , y + y1 + y2 + 2)
These are equal for all (x, y), (x1 , y1 ) and (x2 , y2 ) in V .
A4. (x, y) + (0, −1) = (x + 0, y + (−1) + 1) = (x, y) for all (x, y), so (0, −1) is the zero of V .
A5. (x, y) + (−x, −y − 2) = (x + (−x), y + (−y − 2) + 1) = (0, −1) is the zero of V (from A4) so
the negative of (x, y) is (−x, −y − 2).
S1. a(x, y) = (ax, ay + a − 1) is in V for all (x, y) in V and a in R.
S2. a[(x, y) + (x1 , y1 )] = a(x + x1 , y + y1 + 1) = (a(x + x1 ), a(y + y1 + 1) + a − 1)
= (ax + ax1 , ay + ay1 + 2a − 1)
a(x, y) + a(x1 , y1 ) = (ax, ay + a − 1) + (ax1 , ay1 − a − 1)
= ((ax + ax1 ), (ay + a − 1) + (ay1 + a − 1) + 1)
= (ax + ax1 , ay + ay1 + 2a − 1)
These are equal.
S4. a[b(x, y)] = a(bx, by + b − 1) = (a(bx), a(by + b − 1) + a − 1) = (abx, aby + ab − 1) =
(ab)(x, y).
S5. 1(x, y) = (1x, 1y + 1 − 1) = (x, y) for all (x, y) in V .

5. b. Subtract the first equation from the second to get x − 3y = v − u, whence x = 3y + v − u.

Substitute in the first equation to get

3(3y + v − u) − 2y = u
7y = 4u − 3v
 6.1. Examples and Basic Properties 91

y = 47 u − 73 v

Substitute this in the first equation to get x = 57 u − 27 v.

It is worth noting that these equations can also be solved by gaussian elimination using u and
v as the constants.
h i h i h i h i
6. b. au + bv + cw = 0 becomes a0 0a + 0b b0 + cc −cc = 00 00 .
Equating corresponding entries gives equations for a and b.

a + c = 0, b + c = 0, b + c = 0, a − c = 0

The only solution is a = b = c = 0.

d. au + bv + cw = 0 means a sin x + b cos x + c1 = 0 for all choices of x. If x = 0, π2 , π , we get,
respectively, equations b +c = 0, a +c = 0, and −b +c = 0. The only solution is a = b = c = 0.

7. b. 4(3u − v + w) − 2[(3u − 2v) − 3(v − w)] + 6(w − u − v)

= (12u − 4v + 4w) − 2[3u − 2v − 3v + 3w] + (6w − 6u − 6v)
= (12u − 4v + 4w) − (6u − 10v + 6w) + (6w − 6u − 6v)
= 4w
10. Suppose that a vector z has the property that z + v = v for all v in V . Since 0 + v = v also holds for
all v, we obtain z + v = 0 + v, so z = 0 by cancellation.

12. b. (−a)v+av = (−a+a)v = 0v = 0. Since also −(av)+av = 0 we get (−a)v+av = −(av)+av.

Thus (−a)v = −(av) by cancellation.
Alternatively: (−a)v = [(−1)a]v = (−1)(av) = −av using part 4 of Theorem 6.1.3.

13. b. We proceed by induction on n (see Appendix A). The case n = 1 is clear. If the equation holds
for some n ≥ 1, we have

(a1 + a2 + · · · + an + an+1 )v = [(a1 + a2 + · · · + an ) + an+1 ]v

= (a1 + a2 + · · · + an )v + an+1 v by S3
= (a1 v + a2 v + · · · + an v) + an+1 v by induction
= a1 v + a2 v + · · · + an v + an+1 v

Hence it holds for n + 1, and the induction is complete.

15. c. Since a 6= 0, a−1 exists in R. Hence av = aw gives a−1 av = a−1 aw; that is 1v = 1w, that is
v = w.
Alternatively: av = aw gives av − aw = 0, so a(v − w) = 0. As a 6= 0, it follows that v − w = 0
by Theorem 6.1.3, that is v = w.
92 Vector Spaces

6.2 Subspaces and Spanning Sets

1. b. Yes. U is a subset of P3 because xg(x) has degree one more than the degree of g(x). Clearly
0 = x · 0 is in U . Given u = xg(x) and v = xh(x) in U (where g(x) and b(x) are in P2 ) we have

u + v = x(g(x) + h(x)) is in U because g(x) + h(x) is in P2

ku = x(kg(x)) is in U for all k in R because kg(x) is in P2

d. Yes. As in (b), U is a subset of P3 . Clearly 0 = x · 0 + (1 − x) · 0 is in U . If u = xg(x) + (1 −

x)h(x) and v = xg1 (x) + (1 − x)h1 (x) are in U then

u + v = x[g(x) + g1 (x)] + (1 − x) [h(x) + h1 (x)]

ku = x[kg(x)] + (1 − x)[kh(x)]

both lie in U because g(x) + g1 (x) and h(x) + h1 (x) are in P2 .

f. No. U is not closed under addition (for example u = 1 + x3 and v = x − x3 are in U but
u + v = 1 + x is not in U ). Also, the zero polynomial is not in U .
h i h i h i
0 0 a b a1 b1
2. b. Yes. Clearly 0 = 0 0 is in U . If u = c d and u1 = c1 d1 are in U then u + u1 =
h i
a + a1 b + b1
c + c1 d + d1 is in U because

(a + a1 ) + (b + b1) = (a + b) + (a1 + b1 )
= (c + d) + (c1 + d1)
= (c + c1 ) + (d + d1 )
h i
ka kb
ku = kc kd is in U because ka + kb = k(a + b) = k(c + d) = kc + kd.
d. Yes. Here 0 is in U as 0B = 0. If A and A1 are in U then AB = 0 and A1 B = 0, so (A + A1 )B =
AB + A1 B = 0 + 0 = 0 and (kA)B = k(AB) = k0 = 0 for all k in R. This shows that A + A1 and
kA are also in U .
h i h i
1 0 0 0
f. No. U is not closed under addition. In fact, A = 0 0 and A1 = 0 1 are both in U , but
h i
A + A1 = 10 01 is not in U .

3. b. No. U is not closed under addition. For example if f and g are defined by f (x) = x + 1 and
g(x) = x2 + 1, then f and g are in U but f + g is not in U because ( f + g)(0) = f (0) + g(0) =
1 + 1 = 2.
d. No. U is not closed under scalar multiplication. For example, if f is defined by f (x) = x, then
f is in U but (−1) f is not in U (for example [(−1) f ]( 12 ) = − 21 so is not in U ).
f. Yes. 0 is in U because 0(x + y) = 0 = 0 + 0 = 0(x) + 0(y) for all x and y in [0, 1]. If f and g
are in U then, for all k in R:

( f + g)(x + y) = f (x + y) + g(x + y)
= ( f (x) + f (y)) + (g(x) + g(y))
 6.2. Subspaces and Spanning Sets 93

= ( f (x) + g(x)) + ( f (y) + g(y))

= ( f + g)(x) + ( f + g)(y)

(k f )(x + y) = k[ f (x + y)] = k[ f (x) + f (y)] = k[ f (x)] + k[ f (y)]

= (k f )(x) + (k f )(y)

Hence f + g and k f are in U .

" x # " y1
#
1
.. ..
5. b. Suppose x = . 6= 0, say xk 6= 0. Given y = . let A be the m × n matrix with kth
xn yn
column x−1
k y and the other columns zero. Then y = Ax by matrix multiplication, so y is in U .
Since y was an arbitrary column in Rn , this shows that U = Rm .
6. b. We want r, s and t such that 2x2 −3x+1 = r(x+1)+s(x2 +x)+t(x2 +2). Equating coefficients
of x2 , x and 1 gives s + t = 2, r + s = −3, r + 2t = 1. The unique solution is r = −3, s = 0,
t = 2.
d. As in (b), x = 23 (x + 1) + 13 (x2 + x) − 13 (x2 + 2).

7. b. If v = su + tw then x = s(x2 + 1) + t(x + 2). Equating coefficients gives 0 = s, 1 = t and

0 = s + 2t. Since there is no solution to these equations, v does not lie in span {u, w}.
h i h i h i
d. If v = su + tw, then 15 −43 = s 12 −11 + t 21 10 . Equating corresponding entries gives
s + 2t = 1, −s +t = −4, 2s +t = 5 and s = 3. These equations have the unique solution t = −1,
s = 3, so v is in span {u, w}; in fact v = 3u − w.
b. Yes. The trigonometry identity 1 = sin2 x+cos2 x for all x means that 1 is in span sin2 x, cos2 x .

8.
d. Suppose 1 + x2 = s sin2 x + t cos2 x for some s and t. This must hold for all x. Taking x = 0
gives 1 = t; taking x = π gives 1 + π 2  = −t. Thus 2 + π 2 = 0, a contradiction. So no such s
and t exist, that is 1 + x2 is not in span sin2 x, cos2 x .

b. Write U = span 1 + 2x2 , 3x, 1 + x , then successively


9.

x = 13 (3x) is in U
1 = (1 + x) − x is in U
x2 = 12 [(1 + 2x2 ) − 1] is in U

Since P2 = span 1, x, x2 , this shows that P2 ⊆ U . Clearly U ⊆ P2 , so U = P2 .



11. b. The vectors u−v = 1u+(−1)v, u+v, and w are all in span {u, v, w} so span {u − v, u + w, w} ⊆
span {u, v, w} by Theorem 6.2.2. The other inclusion also follows from Theorem 6.2.2 be-
cause

u = (u + w) − w
v = −(u − v) + (u + w) − w
w=w

show that u, v and w are all in span {u − v, u + v, w}.

94 Vector Spaces

14. No. For example (1, 1, 0) is not even in span {(1, 2, 0), (1, 1, 1)}. Indeed (1, 1, 0) = s(1, 2, 0) +
t(1, 1, 1) requires that s + t = 1, 2s + t = 1, t = 0, and this has no solution.

18. Write W = span {u, v2 , . . . , vn }. Since u is in V we have W ⊆ V . But the fact that a1 6= 0 means
1 a2 an
v1 = a1 u − a1 v2 − · · · − a1 vn

so v1 is in W . Since v2 , . . . , vn are all in W , this shows that V = span {v1 , v2 , . . . , vn } ⊆ W . Hence

V = W.

21. b. If u and u + v are in U then v = (u + v) − u = (u + v) + (−1)u is in U because U is closed

under addition and scalar multiplication.

22. If U is a subspace then, u1 + au2 is in U for any ui in U and a in R by the subspace test. Conversely,
assume that this condition holds for U . Then, in the subspace test, conditions (2) and (3) hold for
U (because 1v = v for all v in V ), so it remains to show that 0 is in U . This is where we use
the assumption that U is nonempty because, if u is any vector in U then u + (−1)u is in U by
assumption, that is 0 ∈ U .

6.3 Linear Independence and Dimension

1. b. Independent. If rx2 + s(x + 1) + t(1 − x − x2 ) = 0 then, equating coefficients of x2 , x and 1, we

get r − t = 0, s − t = 0, s + t = 0. The only solution is r = s = t = 0.
h i h i h i h i h i
d. Independent. If r 11 10 + s 01 11 + t 11 01 + u 10 11 = 00 00 , then r + t + u = 0,
r + s + u = 0, r + s + t = 0, s + t + u = 0. The only solution is r = s = t = u = 0.

2. b. Dependent. 3(x2 − x + 3) − 2(2x2 + x + 5) + (x2 + 5x + 1) = 0

h i h i h i h i h i
d. Dependent. 2 −10 −10 + −11 −11 + 11 11 + 0 −10 −10 = 0
0
0
0 .
5 1
f. Dependent. x2 +x−6
+ x2 −5x+6 − x26−9 = 0.

3. b. Dependent. 1 − sin2 x − cos2 x = 0 for all x.

4. b. If r(2, x, 1) + s(1, 0, 1) + t(0, 1, 3) = (0, 0, 0) then, equating components:

2r + s = 0
xr + t = 0
r + s + 3t = 0

Gaussian elimination gives

       
2 1 0 0 1 1 3 0 1 1 3 0 1 1 3 0
x 0 1 0 → 2 1 0 0 → 0 1 6 0 → 0 1 6 0
1 1 3 0 x 0 1 0 0 −x 1 − 3x 0 0 0 1 + 3x 0
 6.3. Linear Independence and Dimension 95

This has only the trivial solution r = s = t = 0 if and only if x 6= −1

3 . Alternatively, the coeffi-
cient matrix has determinant
   
2 1 0 2 1 0 h i
det x 0 1 = det x 0 1 = − det −1x 13 = −(1 + 3x)
1 1 3 −1 0 3

This is nonzero if and only if x 6= − 31 .

5. b. Independence: If r(−1, 1, 1) + s(1, −1, 1) + t(1, 1, −1) = (0, 0, 0) then −r + s + t = 0,

r − s + t = 0, r + s − t = 0. The only solution is r = s = t = 0.
Spanning: Write U = span {(−1, 1, 1), (1, −1, 1), (1, 1, −1)}.
Then (1, 0, 0) = 12 [(1, 1, −1) + (1, −1, 1)] is in U ; similarly (0, 1, 0) and (0, 0, 1) are in U .
As R3 = span {(1, 0, 0), (0, 1, 0), (0, 0, 1)}, we have R3 ⊆ U . Clearly U ⊆ R3 , so we have
R3 = U .
d. Independence: If r(1 + x) + s(x + x2 ) + t(x2 + x3 ) + ux3 = 0 then

r + (r + s)x + (s + t)x2 + (t + u)x3 = 0

so r = 0, r + s = 0, s + t = 0, t + u = 0. The only solution is r = s = t = u = 0.

Spanning: Write U = span 1 + x, x + x2 , x2 + x3 , x3 . Then x3 is in U ; whence x2 = (x2 +


x3 ) − x3 is in U ; whence x = (x + x2 ) − x2 is in U ; whence 1 = (1 + x) − x is in U . Hence

P3 = span 1, x, x2 , x3 is contained in U . As U ⊆ P3 , we have U = P3 .

b. Write U = a + b(x + x2 ) | a, b in R = span B where B = 1, x + x2 . But B is independent

 
6.
because s + t(x + x2 ) = 0 implies s = t = 0. Hence B is a basis of U , so dim U = 2.
d. Write U = {p(x) | p(x) = p(−x)}. As U ⊆ P2 , write p(x) = a + bx + cx2 be any member
of U . The condition p(x) = p(−x) becomes a  + bx + cx2 = a − bx + cx2 , so b = 0. Thus
U = a + bx2 | a, b in R = span 1, x2 . As 1, x2 is independent (s + tx2 = 0 implies


s = 0 = t), it is a basis of U , so dim U = 2.

n h i h i o h i
7. b. Write U = A | A −11 10 = −11 01 A . If A = xz wy , A is in U if and only if
h ih i h ih i h i h i
x y 1 1 1 1 x y x−y x x+z y+w
z w −1 0 = −1 0 z w , that is z−w z = −x −y .
This holds if and only if x = y + w and z = −y, that is
h i h i h i
+w y 1 1 1 0
A = y−y w = y −1 0 + w 0 1

nh i h io
1 1 1 0
Hence U = span B where B = −1 0 , 0 1 . But B is independent here because
h i h i h i
s −11 10 + t 10 01 = 00 00 means s + t = 0, s = 0, −s = 0, t = 0, so s = t = 0. Thus B
is a basis of U , so dim U = 2.
n h i h i o h i
1 1 0 1 x y
d. Write U = A | A −1 0 = −1 1 A . If A = z w then A is in U if and only if
h ih i h ih i h i h i
x y 1 1 0 1 x y x−y x z w
z w −1 0 = −1 1 z w ; that is z−w z = z−x w−y .
This holds if and only if z = x − y and x = w; that is
h i h i h i
x y 1 0 0 1
A = x − y x = x 1 1 + y −1 0
96 Vector Spaces

nh i h io h i
1 0 0 1 1 0
Thus U = span B where B = 1 1 , −1 0 . But B is independent because s 1 1 +
h i h i
0 1 0 0
t −1 0 = 0 0 implies s = t = 0. Hence B is a basis of U , so dim U = 2.
h i h i h i
x y x+z y+w x y
8. b. If X = z the condition AX = X is
w = 0 0 and this holds if and only
z w
h i h i h i
if z = w = 0. Hence X = 0x 0y = x 10 00 + y 00 10 . So U = span B where B =
nh i h io
1 0 0 1
0 0 , 0 0 . As B is independent, it is a basis of U , so dim U = 2.

10. b. If the
common column sum is m, V has the form 
a q r
V= b p s | a, b, p, q, r, s, m in R = span B where
m−a−b m− p−q m−r−s
             
0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0
B= 0 0 0 , 0 0 0 , 1 0 0 , 0 0 0 , 0 1 0 , 0 0 0 , 0 0 1 .
1 1 1 −1 0 0 −1 0 0 0 −1 0 0 −1 0 0 0 −1 0 0 −1
The set B is independent (a linear combination using coefficients a, b, p, q, r, s, and m yields
the matrix in V , and this is 0 if and only if a = b = p = q = r = s = m = 0). Hence B is a basis
of B, so dim V = 7.

11. b. A general polynomial in P3 has the form p(x) = a + bx + cx2 + dx3 , so

V = (x2 − x)(a + bx + cx2 + dx3 ) | a, b, c, d in R



= a(x2 − x) + bx(x2 − x) + cx2 (x2 − x) + dx3 (x2 − x) | a, b, c, d in R



= span B

where B = (x2 − x), x(x2 − x), x2 (x2 − x), x3 (x2 − x) . We claim that B is independent. For


if a(x2 − x) + bx(x2 − x) + cx2 (x2 − x) + dx3 (x2 − x) = 0 then (a + bx + cx2 + dx3 )(x2 − x) = 0,
whence a + bx + cx2 + dx3 = 0 by the hint in (a). Thus a = b = c = d = 0. [This also follows
by comparing coefficients.] Thus B is a basis of V , so dim V = 4.

12. b. No. If P3 = span { f1 (x), f2 (x), f3 (x), f4 (x)} where fi (0) = 0 for each i, then each polynomial
p(x) in P3 is a linear combination

p(x) = a1 f1 (x) + a2 f2 (x) + a3 f3 (x) + a4 f4 (x)

when the ai are in R. But then

p(0) = a1 f1 (0) + a2 f2 (0) + a3 f3 (0) + a4 f4 (0) = 0

for every p(x) in P3 . This is not the case, so no such basis of P3 can exist. [Indeed, no such
spanning set of P3 can exist.]
nh i h i h i h io
1 0 1 1 1 0 0 1
d. No. B = 0 1 , 0 1 , , is a basis of invertible matrices.
h i h 1 i1 h1 1 i h i h i
Independent: r 10 01 + s 10 11 + t 11 01 + u 01 11 = 00 00 gives
r + s + t = 0, s + u = 0, t + u = 0, r + s + t + u = 0. The only solution is r = s = t = u = 0.
 6.3. Linear Independence and Dimension 97

h i h i h i
0 1 1 1 1 0
Spanning: 0 0 = 0 1 − 0 1 is in span B
h i h i h i
0 0 1 0 1 0
1 0 = 1 1 − 0 1 is in span B
h i h i h i h i
0 0 0 1 0 1 0 0
0 1 = 1 1 − 0 0 − 1 0 is in span B
h i h i
h i
1 0 1 0 0 0
0 0 = 0 1 − 0 1 is in span B
nh i h i h i h io
0 1
Hence M22 = span 0 0 , 01 0
0 , 0
0
0
1 , 1 0
0 0 ⊆ span B. Clearly span B ⊆
M22 .
f. Yes. Indeed, 0u + 0v + 0w = 0 for any u, v, w, independent or not!
h. Yes. If su + t(u + v) = 0 then (s + t)u + tv = 0, so s + t = 0 and t = 0 (because {u, v} is
independent). Thus s = t = 0.
j. Yes. If su + tv = 0 then su + tv + 0w = 0, so s = t = 0 (because {u, v, w} is independent).
This shows that {u, v} is independent.
l. Yes. Since {u, v, w} is independent, the vector u + v + w is not zero. Hence {u + v + w} is
independent (see Example 5.2.5).
n. Yes. If I is a set of independent vectors, then |I| ≤ n by the fundamental theorem because V
contains a spanning set of n vectors (any basis).
15. If a linear combination of the vectors in the subset vanishes, it is a linear combination of the vectors
in the larger set (take the coefficients outside the subset to be zero). Since it still vanishes, all the
coefficients are zero because the larger set is independent.
19. We have su′ + tv = s(au + bv) + t(cu + dv) = (sa + tc)u + (sb + td)v. Since {u, v} is independent,
we have
su′ + tv′ = 0 if and only if sa
h + tc i=h 0 and h + td
i sb i =0
a c s 0
if and only if b d t = 0
h ih i h i h i h i
′ ′ a c
Hence {u , v } is independent if and only if b d s
t = 0 implies st = 00 .
0

By Theorem 2.4.5, this is equivalent to A being invertible.

23. b. Independent: If r(u+v) +s(v+w) +t(w+u) = 0 then (r +t)u+(r +s)v+(s +t)w+0z = 0.
Thus r +t = 0, r + s = 0, s +t = 0 (because {u, v, w, z} is independent). Hence r = s = t = 0.
d. Dependent: (u + v) − (v + w) + (w + z) − (z + u) = 0 is a nontrivial linear combination that
vanishes.
26. If rz + sz2 = 0, r, s in R, then z(r + sz) = 0. If z is not real then z 6= 0 so r + sz = 0. Thus s = 0
(otherwise z = −r 2
s is real), whence r = 0. Conversely, if z is real then rz + sz = 0 when r = z,
s = −1, so z, z2 is not independent.
29. b. If U is not invertible, let U x = 0 where x 6= 0 in Rn (Theorem 2.4.5). We claim that no set
{A1U , A2U , . . . } can span Mmn (let alone be a basis). For if it did, we could write any matrix
B in Mmn as a linear combination
B = a1 A1U + a2 A2U + · · ·
98 Vector Spaces

Then Bx = a1 AU x + a2 A2U x + · · · = 0 + 0 + · · · = 0, a contradiction. In fact, if entry k of x is

nonzero, then Bx 6= 0 where all entries of B are zero except column k, which consists of 1’s.
33. b. Suppose U ∩ W = 0. If su + tw = 0 with u and w nonzero in U and W , then su = −tw is
in U ∩ W = {0}. Hence su = 0 = tw. So s = 0 = t (as u 6= 0 and w 6= 0). Thus {u, v}
is independent. Conversely, assume that the condition holds. If v 6= 0 lies in U ∩ W , then
{v, −v} is independent by the hypothesis, a contradiction because 1v + 1(−v) = 0.
36. b. If p(x) = a0 + a1 x + · · · + an xn is in On , then p(−x) = −p(x), so
a0 − a1 x + a2 x2 + a3 x3 + a4 x4 − · · · = −a0 − a1 x − a2 x2 − a3 x3 − a4 x4 − · · ·
Hence a0 = a2 = a4 = · · · = 0 and p(x) = a1 x+a3 x3 +a5 x5 +· · · . Thus On = span x, x3 , x5 , . . .


is spanned by the odd powers of x in Pn . The set B = x, x3 , x5 , . . . is independent (because

1, x, x2 , x3 , x4 , . . . is independent) so it is a basis of On . If n is even, B = x, x3 , x5 , . . . , xn−1
 

has n2 members, so dim On = n2 . If n is odd, B = x, x3 , x5 , . . . , xn has n+1


2 members, so
dim On = n+1 2 .

6.4 Finite Dimensional Spaces

1. b. B = {(1, 0, 0), (0, 1, 0), (0, 1, 1)} is independent as r(1, 0, 0) + s(0, 1, 0) + t(0, 1, 1) =
(0, 0, 0) implies r = 0, s + t = 0, t = 0, whence r = s = t = 0. Hence B is a basis by Theorem
6.4.3 because dim R3 = 3.
d. B = 1, x, x2 − x + 1 is independent because r1 + sx + t(x2 − x − 1) = 0 implies r − t = 0,


s − t = 0, and t = 0; whence r = s = t = 0. Hence B is a basis by Theorem 6.4.3 because

dim P2 = 3.
2. b. As dim P2 = 3, any independent set of three 2vectors is a basis by Theorem 6.4.3. But we have
2 2 2
−(x + 3) + 2(x + 2) + (x − 2x − 1) = 0, x + 3, x + 2, x − 2x − 1 , so is dependent. How-
ever any other subset of three vectors from x2 + 3, x + 2, x2 − 2x − 1, x2 + x is independent
(verify).
3. b. B = {(0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 1, 1), (1, 1, 1, 1)} spans R4 because
(1, 0, 0, 0) = (1, 1, 1, 1) − (0, 1, 0, 0) − (0, 0, 1, 1) is in span B
(0, 0, 0, 1) = (0, 0, 1, 1) − (0, 0, 1, 0) is in span B
and, of course, (0, 1, 0, 0) and (0, 0, 1, 0) are in span B. Hence B is a basis of R4 by Theorem
6.4.3 because dim R4 = 4.
d. B = 1, x2 + x, x2 + 1, x3 spans P3 because x2 = (x2 + 1) − 1 and x = (x2 + x) − x2 are in


span B (together with 1 and x3 ). So B is a basis of P3 by Theorem 6.4.3 because dim P3 = 4.

4. b. Let z = a + bi; a, b in R. Then b 6= 0 as z is not real and a 6= 0 as z is not pure imaginary. Since
dim C = 2, it suffices (by Theorem 6.4.3) to show that {z, z} is independent. If rz + sz = 0
then 0 = r(a + bi) + s(a − bi) = (r + s)a + (r − s)bi. Hence (r + s)a = 0 = (r − s)b so (because
a 6= 0 6= b) r + s = 0 = r − s. Thus r = s = 0.
 6.4. Finite Dimensional Spaces 99

5. b. The four polynomials in S have distinct degrees. Use Example 6.3.4.

6. b. {4, 4x, 4x2 , 4x3 } is such a basis. There is no basis of P3 consisting of polynomials have the
property that their coefficients sum to zero. For if it did then every polynomial in P3 would have
this property (since sums and scalar multiples of such polynomials have the same property).

7. b. Not a basis because (2u + v + 3w) − (3u + v − w) + (u − 4w) = 0.

d. Not a basis because 2u − (u + w) − (u − w) + 0(v + w) = 0.

8. b. Yes, four vectors can span R3 — say any basis together with any other vector.
No, four vectors in R3 cannot be independent by the fundamental theorem (Theorem 6.3.2)
because R3 is spanned by 3 vectors ( dim R3 = 3).

10. We have det A = 0 if and only if A is not invertible. This holds if and only if the rows of A are
dependent by Theorem 5.2.3. This in turn holds if and only if some row is a linear combination of
the rest by the dependent lemma (Lemma 6.4.3).

11. b. No. Take X = {(0, 1), (1, 0)} and D = {(0, 1), (1, 0), (1, 1)}. Then D is dependent, but its
subset X is independent.
d. Yes. This is follows from Exercise 15 Section 6.3 (solution above).

15. Let {u1 , . . . , um }, m ≤ k, be a basis of U so dim U = m. If v ∈ U then W = U by Theorem 6.2.2,

so certainly dim W = dim U . On the other hand, if v ∈ / U then {u1 , . . . , um , v} is independent by
the independent lemma (Lemma 6.4.1). Since W = span {u1 , . . . , um , v}, again by Theorem 6.2.2,
it is a basis of W and so dim W = 1 + dim U .

18. b. The two-dimensional subspaces of R3 are the planes through the origin, and the one-dimensional
subspaces are the lines through the origin. Hence part (a) asserts that if U and W are distinct
planes through the origin, then U ∩W is a line through the origin.

23. b. Let vn denote the sequence with 1 in the nth coordinate and zeros elsewhere. Thus v0 =
(1, 0, 0, . . . ), v1 = (0, 1, 0, . . . ) etc. Then a0 v0 +a1 v1 +· · ·+an vn = (a0 , a1 , . . . , an , 0, 0, . . . )
so a0 v0 + a1 v1 + · · · + an vn = 0 implies a0 = a1 = · · · = an = 0. Thus {v0 , v1 , . . . , vn } is an
independent set of n+1 vectors. Since n is arbitrary, dim V cannot be finite by the fundamental
theorem.

25. b. Observe that Ru = {su | s in R}. Hence Ru + Rv = {su + tv | s in R, t in R} is the set of all
linear combinations of u and v. But this is the definition of span {u, v}.
100 Vector Spaces

6.5 An Application to Polynomials

2. b. f (0) (x) = f (x) = x3 + x + 1, so f (1) (x) = 3x2 + 1, f (2) (x) = 6x, f (3) (x) = 6. Hence, Taylor’s
theorem gives
f (2) (1) f (3) (1)
f (x) = f (0) (1) + f (1) (1)(x − 1) + 2
2! (x − 1) + 3! (x − 1)
3

= 3 + 4(x − 1) + 3(x − 1)2 + (x − 1) 3

d. f (0) (x) = f (x) = x3 − 3x2 + 3x, f (1) (x) = 3x2 − 6x + 3, f (2) (x) = 6x − 6, f (3) (x) = 6. Hence,
Taylor’s theorem gives
f (2) (1) f (3) (1)
f (x) = f (0) (1) + f (1) (1)(x − 1) + 2
2! (x − 1) + 3! (x − 1)
3

0
= 1 + 0(x − 1) + 2! (x − 1)2 + 1(x − 1) 3

= 1 + (x − 1)3

6. b. The three polynomials are x2 − 3x + 2 = (x − 1)(x − 2), x2 − 4x + 3 = (x − 1)(x − 3) and x2 −

5x + 6 = (x − 2)(x − 3), so use a0 = 3, a1 = 2, a2 = 1, in Theorem 6.5.2.

7. b. The Lagrange polynomials for a0 = 1, a1 = 2, a2 = 3, are

(x−2)(x−3)
δ0 (x) = (1−2)(1−3) = 12 (x − 2)(x − 3)
(x−1)(x−3)
δ1 (x) = (2−1)(2−3) = −(x − 1)(x − 3)
(x−1)(x−2)
δ2 (x) = (3−1)(3−2) = 12 (x − 1)(x − 2)

Given f (x) = x2 + x + 1:

f (x) = f (1)δ0 (x) + f (2)δ1 (x) + f (3)δ2 (x)

= 32 (x − 2)(x − 3) − 7(x − 1)(x − 3) + 13
2 (x − 1)(x − 2)

10. b. If r(x − a)2 + s(x − a)(x − b) + t(x − b)2 = 0, then taking x = a gives t(a − b)2 = 0, so t =
0 because a 6= b; and taking x = b gives r(b − a)2 = 0, so r = 0. Thus, we are left with
s(x −a)(x − b) = 0. If x is any number except a, b, this implies s = 0. Thus
B = (x − a)2 , (x − a)(x − b), (x − b)2 is independent in P2 ; since dim P2 = 3, B is a basis.

11. b. Have Un = { f (x) in Pn | f (a) = 0 = f (b)}. Let {p1 (x), . . . , pn−1 (x)} be a basis of Pn−2 ; it
suffices to show that

B = {(x − a)(x − b)p1 (x), . . . , (x − a)(x − b)pn−1 (x)}

is a basis of Un . Clearly B ⊆ Un .
Independent: Let s1 (x − a)(x − b)p1 (x) + · · · + sn−1 (x − a)(x − b)pn−1 (x) = 0. Then (x −
a)(x − b)[s1 p1 (x) + · · · + sn−1 pn−1 (x)] = 0, so (by the hint) s1 p1 (x) + · · · + sn−1 pn−1 (x) = 0.
Thus s1 = s2 = · · · = sn−1 = 0.
 6.6. An Application to Differential Equations 101

Spanning: Given f (x) in Pn with f (a) = 0, we have f (x) = (x − a)g(x) for some polynomial
g(x) in Pn−1 by the factor theorem. But 0 = f (b) = (b − a)g(b) so (as b 6= a) g(b) = 0. Then
g(x) = (x − b)h(x) with h(x) = r1 p1 (x) + · · · + rn−1 pn−1 (x), ri in R, whence
f (x) = (x − a)g(x)
= (x − a)(x − b)g(x)
= (x − a)(x − b)[r1 p1 (x) + · · · + rn−1 pn−1 (x)]
= r1 (x − a)(x − b)p1 (x) + · · · + rn−1 (x − a)(x − b)pn−1 (x)

6.6 An Application to Differential Equations

1. b. By Theorem 6.6.1, f (x) = ce−x for some constant c. We have 1 = f (1) = ce−1 , so c = e. Thus
f (x) = e1−x .
d. The characteristic polynomial is x2 + x − 6 = (x − 2)(x + 3). Hence f (x) = ce2x + de−3x for
1
some c, d. We have 0 = f (0) = c+d and 1 = f (1) = ce2 +de−3 . Hence, d = −c and c = e2 −e −3
e2x −e−3x
so f (x) = e2 −e−3
.
f. The characteristic polynomial is x2 − 4x + 4 = (x − 2)2 . Hence, f (x) = ce2x + dxe2x = (c +
dx)e2x for some c, d. We have 2 = f (0) = c and 0 = f (−1) = (c − d)e−2 . Thus c = d = 2 and
f (x) = 2(1 + x)e2x .
h. The characteristic polynomial is x2 − a2 = (x − a)(x + a), so (as a 6= −a) f (x) = ceax + de−ax
for some c, d. We have 1 = f (0) = c + d and 0 = f (1) = cea + de−a . Thus d = 1 − c and
1
c = 1−e 2a whence
eax −ea(2−x)
f (x) = cax + (1 − c)e−ax = 1−e2a
j. The characteristic polynomial is x2 + 4x + 5. The roots are λ = −2 ± i, so
f (x) = e−2x (c sin x + d cos x) for some real c and d.
We have 0 = f (0) = d and 1 = f π2 = e−π (c). Hence f (x) = eπ −2x sin x.

4. b. If f (x) = g(x) + 2 then f ′ + f = 2 becomes g′ + g = 0, whence g(x) = ce−x for some c. Thus
f (x) = ce−x + 2 for some constant c.
3
5. b. If f (x) = − x3 then f ′ (x) = −x2 and f ′′ (x) = −2x, so
f ′′ (x) + f ′ (x) − 6 f (x) = −2x − x2 + 2x3
3
Hence, f (x) = −x3 is a particular solution. Now, if h = h(x) is any solution, write g(x) =
3
h(x) − f (x) = h(x) + x3 . Then
g′′ + g′ − 6g = (h′ + h′ − 6h) − ( f ′′ + f ′ − 6 f ) = 0
So, to find g, the characteristic polynomial is x2 + x − 6 = (x − 2)(x + 3). Hence we have
g(x) = ce−3x + de2x , where c and d are constants, so
3
h(x) = ce−3x + de2x − x3
102 Vector Spaces

3 ln(1/2)
6. b. The general solution is m(t) = 10( 45 )1/3 . Hence 10( 45 )t/3 = 5 so t = ln(4/5) = 9.32 hours.

7. b. If m = m(t) is the mass at time t, then the rate m′ (t) of decay is proportional to m(t), that is
m′ (t) = km(t) for some k. Thus, m′ −km = 0 so m = cekt for some constant c. Since m(0) = 10,

1/3
we obtain c = 10, whence m(t) = 10ekt . Also, 8 = m(3) = 10e3k so e3k = 45 , ek = 45 ,
t/3
m(t) = 10(ek )t = 10 45


.
2π 2π
9. In Example 6.6.4, we found that the period of oscillation is √ . Hence √ = 30 so we obtain
k k
π 2


k = 15 = 0.044.

Supplementary Exercises: Chapter 6

2. b. Suppose {Ax1 , . . . , Axn } is a basis of Rn . To show that A is invertible, we show that YA = 0

implies Y = 0. (This shows AT is invertible by Theorem 2.4.4,so A is invertible). So assume
that YA = 0. Let c1 , . . . , cmdenote
 the columns of Im , so  Im = c1 c2 · · · cm . Then Y =
Y Im = Y c1 c2 · · · cm = Y c1 Y c2 · · · Y cm , so it suffices to show that Y c j = 0 for
each j. But c j is in Rn so our hypothesis shows that c j = r1 Av1 + · · · + rn Avn for some r j in R.
Hence,
c j = A(r1 v1 + · · · + rn vn )
so Y c j = YA(r1 v1 + · · · + rn vn ) = 0, as required.

4. Assume that A is m × n. If x is in null A, then Ax = 0 so (AT A)x = AT 0 = 0. Thus x is in null AT A,

so null A ⊆ null AT A. Conversely, let x be in null AT A; that is AT Ax = 0. Write
" y #
1
..
Ax = y = .
ym

Then y21 + y22 + · · · + y2m = yT y = (Ax)T (Ax) = xT AT Ax = xT 0 = 0. Since the yi are real numbers,
this implies that y1 = y2 = · · · = ym = 0; that is y = 0, that is Ax = 0, that is x is in null A.
7. Linear Transformations

7.1 Examples and Elementary Properties

 
1 0 0
1. b. T (X ) = X A where A = 0 1 0 and X = (x, y, z) is thought of as a row matrix. Hence,
0 0 −1
matrix algebra gives T (X +Y ) = A(X +Y ) = AX + AY = T (X ) + T (Y ) and T (rλ ) = A(rX ) =
rA(X ) = rT (X ).
d. T (A + B) = P(A + B)Q = PAQ + PBQ = T (A) + T (B); T (rA) = P(rA)Q = rPAQ = rT (A).
f. Here T [p(x)] = p(0) for all polynomials p(x) in Pn . Thus

T [(p + q)(x)] = T [p(x) + q(x)] = p(0) + q(0) = T [p(x)] + T [q(x)]

T [rp(x)] = rp(0) = r[T p(x)]

h. Here z is fixed in Rn and T (x) = x · z for all x in Rn . We use Theorem 5.3.1:

T (x + y) = (x + y) · z = x · z + y · z = T (x) + T (y)
T (rx) = (rx) · z = r(x · z) = rT (x)

j. If v = (r1 · · · rn ) and w = (s1 · · · sn ) then, v + w = (r1 + s1 · · · rn + sn ). Hence:

T (v + w) = (r1 + s1 )e1 + · · · + (rn + sn )en

= (r1 e1 + · · · + rn en ) + (s1 e1 + · · · + sn en ) = T (v) + T (w)

Similarly, for a in R, we have av = (ar1 · · · arn ) so

T (av) = (ar1)e1 + · · · + (arn )en = a(r1e1 + · · · + rn en ) = aT (v)

     
1 0 0 ... 0 1 0 0 ... 0 2 0 0 ... 0
0 1 0 ... 0 0 −1 0 ... 0 0 0 0 ... 0
0 0 0 ... 0 0 0 0 ... 0 0 0 0 ... 0
     
2. b. Let A = 
 .. .. .. .. ..
, B = 
  .. .. .. .. ..
, then A + B = 
  .. .. .. .. ..
. Thus,

. . . . . . . . . . . . . . .
0 0 0 ... 0 0 0 0 ... 0 0 0 0 ... 0
T (A) = rank A = 2, T (B) = rank B = 2 and T (A + B) = rank (A + B) = 1. Thus T (A + B) 6=
T (A) + T (B).
d. Here T (v) = v + u, T (w) = w + u, and T (v + w) = v + w + u. Thus if T (v + w) = T (v) +
T (w) then v + w + u = (v + u) + (w + u), so u = 2u, u = 0. This is contrary to assumption.
Alternatively, T (0) = 0 + u 6= 0, so T cannot be linear by Theorem 7.1.1.

3. b. Because T is linear, T (3v1 + 2v2 ) = 3T (v1 ) + 2T (v2 ) = 3(2) + 2(−3) = 0.

103
104 Linear Transformations

h i h i h i
1 1 1
d. Since we know the action of T on −1 and 1 , it suffices to express −7 as a linear
combination of these vectors.
h i h i h i
1 1 1
−7 =r −1 +s 1

Comparing components gives 1 = r + s and −7 = −r + s. The solution is r = 4, s = −3, so

h i  h i h i h i h i h i h i h i
1 1 1 1 1 0 1 −3
T −7 = T 4 −1 − 3 1 = 4T −1 − 3T 1 = 4 1 − 3 1 = 4

f. We know T (1), T (x + 2) and T (x2 + x), so we express 2 − x + 3x2 as a linear combination of

these vectors:
2 − x + 3x2 = r · 1 + s(x + 2) + t(x2 + x)
Equating coefficients gives 2 = r + 2s, −1 = s + t and 3 = t. The solution is r = 10, s = −4
and t = 3, so
T (2 − x + 3x2 ) = T [r · 1 + s(x + 2) + t(x2 + x)]
= rT (1) + sT (x + 2) + tT (x2 + x)
= 5r + s + 0
= 46

In fact, we can find the action of T on any vector a + bx + cx2 in the same way. Observe that
a + bx + cx2 = (a − 2b + 2c) · 1 + (b − c)(x + 2) + c(x2 + x)
for any a, b and c, so
T (a + bx + cx2 ) = (a − 2b + 2c)T (1) + (b − c)T (x + 2) + cT (x2 + x)
= (a − 2b + 2c) · 5 + (b − c) · 1 + c · 0
= 5a − 9b + 9c

This retrieves the above result when a = 2, b = −1 and c = 3.

4. b. Since B = {(2, −1), (1, 1)} is a basis of R2 , any vector (x, y) in R2 is a linear combination
(x, y) = r(2, −1) + s(1, 1). Indeed, equating components gives x = 2r + s and y = −r + s so
r = 13 (x − y), s = 13 (x + 2y). Hence,
T (x, y) = T [r(2, −1) + s(1, 1)]
= rT (2, −1) + sT (1, 1)
= 31 (x − y)(1, −1, 1) + 13 (x + 2y)(0, 1, 0)
= 13 (x − y), y, 13 (x − y)

= 13 (x − y, 3y, x − y)

In particular, T (v) = T (−1, 2) = 13 (−3, 6, −3) = (−1, 2, −1).

This works in general. Observe that (x, y) = x−y x+2y
3 (2, −1) + 3 (1, 1) for any x and y, so since
T is linear,
T (x, y) = x−y x+2y
3 T (2, −1) + 3 T (1, 1)
for any choice of T (2, −1) and T (1, 1).
 7.1. Examples and Elementary Properties 105

nh i h i h i h io h i
1 0
d. Since B = 0 0 , 01 1
0 , 1
1
0
0 , 0
0
0
1 is a basis of M22 , every vector a
c
b
d is
a linear combination
h i h i h i h i h i
a b 1 0 0 1 1 0 0 0
c d =r 0 0 +s 1 0 +t 1 0 +u 0 1

Indeed, equating components and solving for r, s, t and u gives r = a − c + b, s = b, t = c − b,

u = d. Thus,
h i h i h i h i h i
T ac db = rT 10 00 + sT 01 10 + tT 11 00 + uT 00 01
= (a − c + b) · 3 + b · (−1) + (c − b) · 0 + d · 0
= 3a + 2b − 3c

5. b. Since T is linear, the given conditions read

T (v) + 2T (w) = 3v − w
T (v) − T (w) = 2v − 4w
Add twice the second equation to the first to get 3T (v) = 7v − 9w, T (v) = 37 v − 3w. Similarly,
subtracting the second from the first gives 3T (w) = v + 3w, T (w) = 13 v + w. [Alternatively,
we can use gaussian elimination with constants 3v − w and 2v − 4w.]
8. b. Since {v1 , . . . , vn } is a basis of V , every vector v in V is a unique linear combination v =
r1 v1 + · · · + rn vn , ri in R. Hence, as T is linear,
T (v) = r1 T (v1 ) + · · · + rn T (vn ) = r1 (−v1 ) + · · · + rn (−vn ) = −v = (−1)v
Since this holds for every v in V , it shows that T = −1, the scalar operator.
12. {1} is a basis of the vector space R. If T : R → V is a linear transformation, write T (1) = v. Then,
for all r in R :
T (r) = T (r · 1) = rT (1) = rv
Since T (r) = rv is linear for each v in V , this shows that every linear transformation T : R → V
arises in this way.
15. b. Write U = {v ∈ V | T (v) ∈ P}. If v and v1 are in U , then T (v) and T (v1 ) are in P. As P is
a subspace, it follows that T (v + v1 ) = T (v) + T (v1 ) and T (rv) = rT (v) are both in P; that
is v + v1 and rv are in U . Since 0 is in U —because T (0) = 0 is in P—it follows that U is a
subspace.
18. Assume that {v, T (v)} is independent. Then T (v) 6= v (or else 1v + (−1)T (v) = 0) and similarly
T (v) 6= −v.
Conversely, assume that T (v) 6= v and T (v) 6= −v. To verify that {v, T (v)} is independent, let
rv + sT (v) = 0; we must show that r = s = 0. If s 6= 0, then T (v) = av where a = − rs . Hence
v = T [T (v)] = T (av) = aT (v) = a2 v. Since v 6= 0, this gives a = ±1, contrary to hypothesis. So
s = 0, whence rv = 0 and r = 0.
21. b. Suppose that T : Pn → R is a linear transformation such that T (xk ) = T (x)k holds for all k ≥
0 (where x0 = 1). Write T (x) = a. We have T (xk ) = T(x)k = ak = Ea (xk ) for each k by
assumption. This gives T = Ea by Theorem 7.1.2 because 1, x, x2 , . . . , xi , . . . , xn is a basis
of Pn .
106 Linear Transformations

7.2 Kernel and Image of a Linear Transformation

1. b. We have ker TA = {x | Ax = 0}; to determine this space, we use gaussian elimination:

     
2 1 −1 3 0 1 0 3 1 0 1 0 3 1 0
1 0 3 1 0 → 0 1 −7 1 0 → 0 1 −7 1 0
1 1 −4 2 0 0 1 −7 1 0 0 0 0 0 0

−3s − t −3 1
(" # ) (" # " #)
7s − t 7 1
Hence ker TA = s s, t in R = span 1 , 0 . These vectors are inde-
t 0 −1
pendent so nullity of TA = dim ( ker TA ) = 2. Next
im TA = Ax | x in R4

( " r # )
2 1 −1 3
s
= 1 0 3 1
t r, s, t, u in R
1 1 −4 2
y
         
2 1 −1 3
= r 1 +s 0 +t 3 +u 1 r, s, t, u in R
1 1 −4 2

Thus im TA = col A as is true in general. Hence dim ( im TA ) = dim ( col A) = rank A, and we
can compute this by carrying A to row-echelon form:
   
2 1 −1 3 1 0 3 1
1 0 3 1 → 0 1 −7 1
1 1 −4 2 0 0 0 0

Thus dim ( im TA ) = rank A = 2. However, we want a basis of col A, and we obtain this by
writing the columns of A as rows and carrying the resulting matrix (it is AT ) to row-echelon
form: " 2 1 1 # " 1 0 1 # " 1 0 1 #
1 0 1 0 1 −1 0 1 −1
−1 3 −4 → 0 3 −3 → 0 0 0
3 1 2 0 1 −1 0 0 0
   
1 0
Hence, by Lemma 5.4.2, 0 , 1 is a basis of im TA = col A. Of course this once
1 −1
again shows that rank TA = dim ( col A) = 2.
d. ker TA = {x | Ax = 0} so, as in (b), we use gaussian elimination:
" 2 1 0 # " 1 −1 3 # " 1 0 1
#
1 −1 3 0 3 −6 0 1 −2
1 2 −3 → 0 3 −6 → 0 0 0
0 3 −6 0 3 −6 0 0 0
    
−t −1
Hence, ker TA = 2t t in R = span 2 . Thus the nullity of TA is dim ( ker TA ) =
t 1
1. As in (b), im TA = col A and we find a basis by doing gaussian elimination on AT :
     
2 1 1 0 1 −1 2 3 1 0 1 1
1 −1 2 3 → 0 3 −3 −6 → 0 1 −1 −2
0 3 −3 −6 0 3 −3 −6 0 0 0 0

1 0
(" # " #)
0 1
Hence, im TA = col A = span 1 , −1 , so rank TA = dim ( im TA ) = 2.
1 −2
 7.2. Kernel and Image of a Linear Transformation 107

b. Here T = P2 → R2 given by T (p(x)) =

 
2. p(0) p(1) . Hence
ker T = {p(x) | p(0) = p(1) = 0}
If p(x) = a + bx + cx2 is in ker T , then 0 = p(0) = a and 
0 = p(1) = a + b + c. This means
that p(x) = bx − bx , and so ker T = span x − x . Thus x − x2 is a basis of ker T . Next,
2 2

im T is a subspace of R2 . We have (1, 0) = T (1 − x) and (0, 1) = T (x) are both in im T , so

im T = R2 . Thus {(1, 0), (0, 1)} is a basis of im T .
d. Here T : R3 → R4 given by T (x, y, z) = (x, x, y, y). Thus,
ker T = {(x, y, z) | (x, x, y, y) = (0, 0, 0, 0)} = {(0, 0, z) | z in R} = span {(0, 0, 1)}
Hence, {(0, 0, 1)} is a basis of ker T . On the other hand,
im T = {(x, x, y, y) | x, y in R} = span {(1, 1, 0, 0) , (0, 0, 1, 1)}
Then {(1, 1, 0, 0) , (0, 0, 1, 1)} is a basis of im T .
h i
f. Here T : M22 → R is given by T ac db = a + d. Hence
nh i o nh i o
a b a b
ker T = c d a+d = 0 = c −a a, b, c in R
nh i h i h io
1 0 0 1 0 0
= span 0 −1 , 0 0 , 1 0
nh i h i h io
1 0 0 1 0 0
Hence, 0 −1 , 0 0 , 1 0 is a basis of ker T (being independent). On the other
hand, n h i o
a b
im T = a + d c d in M22 = R
So {1} is a basis of im T .
h. T : Rn → R, T (r1 , r2 , . . . , rn ) = r1 + r2 + · · · + rn . Hence,
ker T = {(r1 , r2 , . . . , rn ) | r1 + r2 + · · · + rn = 0}
= {(r1 , r2 , . . . , rn−1 , −r1 − · · · − rn−1 ) | ri in R}
= span {(1, 0, 0, . . . , −1) , (0, 1, 0, . . . , −1) , . . . , (0, 0, 1, . . . , −1)}
This is a basis of ker T . On the other hand,
im T = {r1 + · · · + rn | (r1 , r2 , . . . , rn ) is in Rn } = R
Thus {1} is a basis of im T .
h i h i
j. T : M22 → M22 is given by T (X ) = X A where A = 10 10 . Writing X = xz wy :
nh i h i h io nh i o
x y x x 0 0 0 y
ker T = {X | X A = 0} = z w z x = 0 0 = 0 w y, w in R
nh i h io
0 1
= span 0 0 , 00 01
nh i h io
0 1 0 0
Thus, 0 0 , 0 1 is a basis of ker T (being independent). On the other hand,
nh i o nh i h io
x x 1 1 0 0
im T = {X A | X in M22 } = z z x, z in R = span 0 0 , 1 1
nh i h io
1 1 0 0
Thus, 0 0 , 1 1 is a basis of im T .
108 Linear Transformations

3. b. We have T : V → R2 given by T (v) = (P(v), Q(v)) where P : V → R and Q : V → R are linear

transformations. T is linear by (a). Now

ker T = {v | T (v) = (0, 0)}

= {v | P(v) = 0 and Q(v) = 0}
= {v | P(v) = 0} ∩ {v | Q(v) = 0}
= ker P ∩ ker Q

4. b. ker T = {(x, y, z) | x + y + z = 0, 2x − y + 3z = 0, z − 3y = 0, 3x + 4z = 0}. Solving:

4
1 1 1 0 1 1 1 0 1 0 0
" # " # " #
3
2 −1 3 0 0 −3 1 0 0 1 − 31 0
0 −3 1 0 → 0 −3 1 0 → 0 0 0 0
3 0 4 0 0 −3 1 0 0 0 0 0

Hence, ker T = {(−4t, t, 3t) | t in R} = span {(−4, 1, 3)}. Hence,

{(1, 0, 0), (0, 1, 0), (−4, 1, 3)} is one basis of R3 containing a basis of ker T . Thus

{T (1, 0, 0), T (0, 1, 0)} = {(1, 2, 0, 3), (1, −1, −3, 0)}

is a basis of im T by Theorem 7.2.5.

6. b. Yes. dim ( im T ) = dim V − dim ( ker T ) = 5 − 2 = 3. As dim W = 3 and im T is a 3-

dimensional subspace, im T = W . Thus, T is onto.
d. No. If ker T = V then T (v) = 0 for all v in V , so T = 0 is the zero transformation. But W need
not be the zero space. For example, T : R2 → R2 defined by T (x, y) = (0, 0) for all (x, y) in
R2 .
f. No. Let T : R2 → R2 be defined by T (x, y) = (y, 0) for all (x, y) ∈ R2 . Then ker T = {(x, 0) |
x ∈ R} = im T .
h. Yes. We always have dim ( im T ) ≤ dim W (because im T is a subspace of W ). Since
dim ( ker T ) ≤ dim W also holds in this case:

dim V = dim ( ker T ) + dim ( im T ) ≤ dim W + dim W = 2 dim W

1
Hence dim W ≥ 2 dim V .
j. No. T : R2 → R2 given by T (x, y) = (x, 0) is not one-to-one (because ker T = {(0, y) | y ∈ R}
is not 0).
l. No. T : R2 → R2 given by T (x, y) = (x, 0) is not onto.
n. No. Define T : R2 → R2 by T (x, y) = (x, 0), and let v1 = (1, 0) and v2 = (0, 1). Then {v1 , v2 }
spans R2 , but {T (v1 ), T (v2 )} = {v1 , 0} does not span R2 .

7. b. Given w in W , we must show that it is a linear combination of T (v1 ), . . . , T (vn ). As T is onto,

w = T (v) for some v in V . Since V = span {v1 , . . . , vn } we can write v = r1 v1 + · · · + rn vn
where each ri is in R. Hence

w = T (v) = T (r1 v1 + · · · + rn vn ) = r1 T (v1 ) + · · · + rn T (vn )

 7.2. Kernel and Image of a Linear Transformation 109

8. b. If T is onto, let v be any vector in V . Then v = T (r1 , . . . , rn ) for some (r1 , . . . , rn ) in Rn ; that
is v = r1 v1 + · · · + rn vn is in span {v1 , . . . , vn }. Thus V = span {v1 , . . . , vn }. Conversely, if
V = span {v1 , . . . , vn }, let v be any vector in V . Then v is in span {v1 , . . . , vn } so r1 , . . . , rn
exist in R such that
v = r1 v1 + · · · + rn vn = T (r1 , . . . , rn )
Thus T is onto.

10. The trace map T : M22 → R is linear (Example 7.1.2) and it is onto (for example,
r = tr [ diag (r, 0, . . . , 0)] = T [ diag (r, 0, . . . , 0)] for any r in R). Hence the dimension theorem
gives dim ( ker T ) = dim Mnn − dim ( im T ) = n2 − dim (R) = n2 − 1.
12. Define TA : Rn → Rm and TB : Rn → Rk by TA (x) = Ax and TB (x) = Bx for all x in Rn . Then the
given condition means ker TA ⊆ ker TB , so dim ( ker TA ) ≤ dim ( ker TB ). Hence

rank A = dim ( im TA ) = n − dim ( ker TA ) ≥ n − dim ( ker TB ) = dim ( im TB ) = rank B

b. Write B = x − 1, x2 − 1, . . . , xn − 1 . Then B ⊆ ker T because T (xk − 1) = 1 − 1 = 0 for


15.
all k. Hence span B ⊆ ker T . Moreover, the polynomials in B are independent (they have
distinct degrees), so dim ( span B) = n. Hence, by Theorem 6.4.2, it suffices to show that
dim ( ker T ) = n. But T : Pn → R is onto, so the dimension theorem gives dim ( ker T ) =
dim (Pn ) − dim (R) = (n + 1) − 1 = n, as required.

20. If we can find an onto linear transformation T : Mnn → Mnn with ker T = U and im T = V , then
we are done by the dimension theorem. The condition ker T = U suggests that we define T by
T (A) = A − AT for all A in Mnn . By Example 7.2.3, T is linear, ker T = U , and im T = V . This is
what we wanted.

22. Fix a column y 6= 0 in Rn , and define T : Mmn → Rm by T (A) = Ay for all A in Mmn . This is linear
and ker T = U , so the dimension theorem gives

mn = dim (Mmn ) = dim ( ker T ) + dim ( im T ) = dim U + dim ( im T )

Hence, it suffices to show that dim ( im T ) = m, equivalently (since im T ⊆ Rm ) that T is onto. So

let x be a column in Rm , we must find a matrix A in Mmn such that Ay = x. Write A in terms of its
   T
columns as A = C1 C2 · · · Cn and write y = y1 y2 · · · yn . Then the requirement that
Ay = x becomes
 y1 
  y2
x = C1 C2 · · · Cn  ..  = y1C1 + y2C2 + · · · + ynCn (∗)
.
yn

Since y 6= 0, let yk 6= 0. Then Ay = x if we choose Ck = y−1

k x and C j = 0 if j 6= k. Hence T is onto
as required.

29. b. Choose a basis {u1 , . . . , um } of U and (by Theorem 6.4.1) let {u1 , . . . , um , . . . , un } be a basis
of V . By Theorem 7.1.3, there is a linear transformation S : V → V such that

S(ui ) = ui if 1 ≤ i ≤ m
S(ui ) = 0 if i > m
110 Linear Transformations

Hence, ui is in im S for 1 ≤ i ≤ m, whence U ⊆ im S. On the other hand, if w is in im S, write

w = S(v), v in V . Then ri exist in R such that

v = r1 u1 + · · · + rm um + · · · + rn un

so

w = r1 S(u1 ) + · · · + rm S(um ) + · · · + rn S(un )

= r1 u1 + · · · + rm um + 0

It follows that w is in U , so im S ⊆ U . Then U = im S as required.

7.3 Isomorphisms and Composition

1. b. T is one-to-one because T (x, y, z) = (0, 0, 0) means x = 0, x +y = 0 and x +y+z = 0, whence

x = y = z = 0. Now T is onto by Theorem 7.3.3.
Alternatively: {T (1, 0, 0), T (0, 1, 0), T (0, 0, 1)} = {(1, 1, 1), (0, 1, 1), (0, 0, 1)} is inde-
pendent, so T is an isomorphism by Theorem 7.3.1.
d. T is one-to-one because T (X ) = 0 implies U XV = 0, whence X = 0 (as U and V are invertible).
Now Theorem 7.3.3 implies that T is onto and so is an isomorphism.
f. T is one-to-one because T (v) = 0 implies kv = 0, so v = 0 because k 6= 0. Hence, T is onto if
dim V is finite (by Theorem 7.3.3) and so is an isomorphism. Alternatively, T is onto because
T (k−1 v) = k(k−1 v) = v holds for all v in V .
h. T is onto because T (AT ) = (AT )T = A for every n × m matrix A (note that AT is in Mmn so
T (AT ) makes sense). Since dim Mmn = mn = dim Mnm , it follows that T is one-to-one by
Theorem 7.3.3, and so is an isomorphism. (A direct proof that T is one-to-one: T (A) = 0
implies AT = 0, whence A = 0.)

4. b. ST (x, y, z) = S(x + y, 0, y + z) = (x + y, 0, y + z); T S(x, y, z) = T (x, 0, z) = (x, 0, z). These

are not equal (if y 6= 0) so ST 6= T S.
h i h i h i h i h i h i
a b c a c 0 a b a 0 0 a
d. ST c d = S d b = 0 b ; T S c d = T 0 d = d 0 . These are not equal
for some values of a, b, c and d (nearly all) so ST 6= T S.

5. b. T 2 (x, y) = T [T (x, y)] = T (x + y, 0) = (x + y + 0, 0) = (x + y, 0) = T (x, y). This holds for all

(x, y), whence T 2 = T .
h i  h i  h i h i
d. T 2 ac db = T T ac db = T 12 aa + c b+d
+c b+d = 1
2 T a+c b+d
a+c b+d
h i h i h i
1 (a + c) + (a + c) (b + d) + (b + d) 1 a+c b+d
= 4 (a + c) + (a + c) (b + d) + (b + d) = 2 a + c b + d = T ac db
h i
a b
This holds for all c d , so T 2 = T .

6. b. No inverse. For example T (1, −1, 1, −1) = (0, 0, 0, 0) so (1, −1, 1, −1) is a nonzero vector
in ker T . Hence T is not one-to-one, and so has no inverse.
 7.3. Isomorphisms and Composition 111

h i h i
d. T is one-to-one because T ac db = 00 00 implies a + 2c = 0 = 3c − a and b + 2d = 0 =
3d − b, whence a = b = c = d = 0. Thus T is an isomorphism by Theorem 7.3.3. If
h i h i h i h i h i
T −1V ac db = xz wy , then ac db = T xz wy = x3z+−2zx 3w y + 2w
− y . Thus

x + 2z = a
−x + 3z = c
y + 2w = b
−y + 3w = d

The solution is x = 15 (3a − 2c), z = 15 (a + c), y = 15 (3b − 2d), w = 15 (b + d). Hence

h i h i
−1 a b 1 3a − 2c 3b − 2d
T c d = 5 a+c b+d (∗)
h i
A better way to find T −1 is to observe that T (X ) = AX where A = −11 23 . This matrix is
invertible which easily implies that T is one-to-one (and onto), and if S : M22 → M22 is defined
by S(X ) = −1 −1
h A X ithen ST = 1M22 and T S = 1M22 . Hence S = T by Theorem 7.3.5. Note that
1 3 −2
A−1 = 5 1 1 which gives (∗).
f. T is one-to-one because, if p in P2 satisfies T (p) = 0, then p(0) = p(1) = p(−1) = 0. If
p = a + bx + cx2 , this means a = 0, a + b + c = 0 and a − b + c = 0, whence a = b = c = 0,
and p = 0. Hence, T −1 exists by Theorem 7.3.3. If T −1 (a, b, c) = r + sx + tx2 , then

(a, b, c) = T (r + sx + tx2 ) = (r, r + s + t, r − s + t)

Then r = a, r + s + t = b, r − s + t = c, whence r = a, s = 21 (b − c), t = 21 (b + c − 2a). Finally

T −1 (a, b, c) = a + 12 (b − c)x + 12 (b + c − 2a)x2

7. b. T 2 (x, y) = T [T (x, y)] = T (ky − x, y) = (ky − (ky − x), y) = (x, y) = 1R2 (x, y). Since this
holds for all (x, y) in R2 , it shows that T 2 = 1R2 . This means that T −1 = T .
d. It is a routine verification that A2 = I. Hence

T 2 (x) = T [T (x)] = A[Ax] = A2 x = Ix = x = 1M22 (x)

holds for all x in M22 . This means that T 2 = 1M22 , and hence that T −1 = T .

8. b. T 2 (x, y, z, w) = T [T (x, y, z, w)] = T (−y, x − y, z, −w)

= (−(x
 2− y), −y − (x  − y), z, −(−w)) = (y − x, −x, z, w)
3
T (x, y, z, w) = T T  3(x, y, z, w) = T (y3 − x, −x, z, w) = (x, y, z, −w)
T 6 (x, y, z, w) = 3
T T (x, y, z, w) = T [x, y, z, −w] = (x, y, z, w) = 1R4 (x, y, z, w)
Hence, T 6 = 1R4 so T −1 = T 5 . Explicitly:

T −1 (x, y, z, w) = T 2 T 3 (x, y, z, w) = T 2 (x, y, z, −w) = (y − x, −x, z, −w)

 
112 Linear Transformations

9. b. Define S : Mnn → Mnn by S(A) = U −1 A. Then

ST (A) = S(T (A)) = U −1 (UA) = A = 1Mnn (A) so ST = 1Mnn

T S(A) = T (S(A)) = U (U −1A) = A = 1Mnn (A) so T S = 1Mnn

Hence, T is invertible and T −1 = S.

T S
10. b. Given V → W → U with T and S both onto, we are to show that ST : V → U is onto. Given
u in U , we have u = S(w) for some w in W because S is onto; then w = T (v) for some v in V
because T is onto. Hence,
ST (v) = S[T (v)] = S[w] = u
This shows that ST is onto.

12. b. If u lies in im RT write u = RT (v), v in V . Thus u = R[T (v)], where T (v) in W , so u is in

im R.
T S
13. b. Given V → U → W with ST onto, let w be a vector in W . Then w = ST (v) for some v in V
because ST is onto, whence w = S[T (v)] where T (v) is in U . This shows that S is onto. Now
the dimension theorem applied to S gives

dim U = dim ( ker S) + dim ( im S) = dim ( ker S) + dim W

because im S = W (S is onto). As dim ( ker S) ≥ 0, this gives dim U ≥ dim W .

14. If T 2 = 1V then T T = 1V so T is invertible and T −1 = T by the definition of the inverse of a

transformation. Conversely, if T −1 = T then T 2 = T T −1 = 1V .

16. Theorem 7.2.5 shows that {T (e1 ), T (e2 ), . . . , T (er )} is a basis of im T . Write
U = span {e1 , . . . , er }. Then B = {e1 , . . . , er } is a basis of U , and T : U → im T carries B to the
basis {T (e1 ), . . . , T (er )}. Thus T : U → im T is itself an isomorphism. Note that T : V → W may
not be an isomorphism, but restricting T to the subspace U of V does result in an isomorphism in
this case.

19. b. We have V = {(x, y) | x, y in R} with a new addition and scalar multiplication:

(x, y) ⊕ (x1 , y1 ) = (x + x1 , y + y1 + 1)
a ⊙ (x, y) = (ax, ay + a − 1)

We use the notation ⊕ and ⊙ for clarity. Define

T : V → R2 by T (x, y) = (x, y + 1)

Then T is a linear transformation because:

T [(x, y) ⊕ (x1 , y1 )] = T (x + x1 , y + y1 + 1)
= (x + x1 , (y + y1 + 1) + 1)
= (x, y + 1) + (x1 , y1 + 1)
= T (x, y) + T (x1 , y1 )
 7.3. Isomorphisms and Composition 113

T (a ⊙ (x, y)] = T (ax, ay + a − 1)

= (ax, ay + a)
= a(x, y + 1)
= aT (x, y)

Moreover T is one-to-one because T (x, y) = (0, 0) means x = 0 = y + 1, so (x, y) = (0, 1), the
zero vector of V . (Alternatively, T (x, y) = T (x1 , y1 ) implies (x, y + 1) = (x1 , y1 + 1), whence
x = x1 , y = y1 .) As T is clearly onto R2 , it is an isomorphism.

24. b. T S[x0, x1 , . . . ) = T [0, x0 , x1 , . . . ) = [x0 , x1 , . . . ) so T S = 1V . Hence T S is both onto and

one-to-one, so T is onto and S is one-to-one by Exercise 10. But [1, 0, 0, . . . ) is in ker T while
[1, 0, 0, . . . ) is not in im S.

26. b. If p(x) is in ker T , then p(x) = −xp′ (x). If we write p(x) = a0 + a1 x + · · · + an xn , this becomes

a0 + a1 x + · · · + an−1 xn−1 + an xn = −a1 x − 2a2 x2 − · · · − nan xn

Equating coefficients gives a0 = 0, a1 = −a1 , a2 = −2a2 , . . . , an = −nan . Hence we have,

a0 = a1 = · · · = an = 0, so p(x) = 0. Thus ker T = {0}, so T is one-to-one. As T : Pn → Pn
and dim Pn is finite, this implies that T is also onto, and so is an isomorphism.

27. b. If T S = 1W then, given w in W , T [S(w)] = w, so T is onto. Conversely, if T is onto, choose a

basis {e1 , . . . , er , er+1 , . . . , en } of V such that {er+1 , . . . , en } is a basis of ker T . By Theorem
7.2.5, {T (e1 ), . . . , T (en )} is a basis of im T = W (as T is onto). Hence, a linear transformation
S : W → V exists such that S[T (ei )] = ei for i = 1, 2, . . . , r. We claim that T S = 1W , and we
show this by verifying that these transformations agree on the basis {T (e1 ), . . . , T (er )} of W .
Indeed
T S[T (ei )] = T {S[T (ei )]} = T (ei ) = 1W [T (ei )]
for i = 1, 2, . . . , n.

28. b. If T = SR, then every vector T (v) in im T has the form T (v) = S[R(v)], whence im T ⊆ im S.
Since R is invertible, S = T R−1 implies im S ⊆ im T , so im S = im T .
Conversely, assume that im S = im T . The dimension theorem gives

dim ( ker S) = n − dim ( im S) = n − dim ( im T ) = dim ( ker T )

Hence, let {e1 , . . . , er , . . . , en } and {f1 , . . . , fr , . . . , fn } be bases of V such that {er+1 , . . . , en }

and {fr+1 , . . . , fn } are bases of ker S and ker T , respectively. By Theorem 7.2.5,
{S(e1 ), . . . , S(er )} and {T (f1 ), . . . , T (fr )} are both bases of im S = im T . So let g1 , . . . , gr
in V be such that
S(ei ) = T (gi )
for each i = 1, 2, . . . , r.
Claim: B = {g1 , . . . , gr , fr+1 , . . . , fn } is a basis of V .
Proof. It suffices (by Theorem 6.4.4) to show that B is independent. If

a1 g1 + · · · + ar gr + br+1 fr+1 + · · · + bn fn = 0
114 Linear Transformations

apply T to get

0 = a1 T (g1 ) + · · · + ar T (gr ) + br+1 T (fr+1 ) + · · · + bn T (fn )

= a1 T (g1 ) + · · · + ar T (gr ) + 0

because T (f j ) = 0 if j˙ > r. Hence a1 = · · · = ar = 0; whence 0 = br+1 fr+1 + · · · + bn fn . This

gives br+1 = · · · = bn = 0 and so proves the claim.
By the claim, we can define R : V → V by

R(gi ) = ei for i = 1, 2, . . . , r
R(f j ) = e j for j = r + 1, . . . , n

Then R is an isomorphism by Theorem 7.3.1, and we claim that SR = T . We show this by

verifying that SR and T have the same effect on the basis B in the claim. The definition of R
gives
SR(gi ) = S[R(gi )] = S(ei ) = T (gi ) for i = 1, 2, . . . , r
SR(f j ) = S[e j ] = 0 = T (f j ) for j = r + 1, . . . , n
Hence SR = T .

29. As in the hint, let {e1 , e2 , . . . , er , . . . , en } be a basis of V where {er+1 , . . . , en } is a basis of

ker T . Then {T (e1 ), . . . , T (er )} is linearly independent by Theorem 7.2.5, so extend it to a ba-
sis {T (e1 ), . . . , T (er ), wr+1 , . . . , wn } of V . Then define S : V → V by

S[T (ei )] = ei for 1 ≤ i ≤ r

S(w j ) = e j for r < j ≤ n

Then, S is an isomorphism (by Theorem 7.3.1) and we claim that T ST = T . We verify this by
showing that T ST and T agree on the basis {e1 , . . . , er , . . . , en } of V (and invoking Theorem 7.1.2).

If 1 ≤ i ≤ r: T ST (ei ) = T (S[T (ei )]) = T (ei )

If r + 1 ≤ j ≤ n: T ST (e j ) = T S[T (e j )] = T S[0] = 0 = T (e j )

where, at the end, we use the fact that e j is in ker T for r + 1 ≤ j ≤ n.

7.4 A Theorem about Differential Equations

This section contains no exercises.

 7.5. More on Linear Recurrences 115

7.5 More on Linear Recurrences

1. b. The associated polynomial is

p(x) = x3 − 7x + 6 = (x − 1)(x − 2)(x + 3)

Hence, {[1), [2n ), [(−3)n )} is a basis of the space of all solutions to the recurrence. The
general solution is thus,
[xn ) = a[1) + b[2n) + c[(−3)n)
where a, b and c are constants. The requirement that x0 = 1, x1 = 2, x2 = 1 determines a, b,
and c. We have
xn = a + b2n + c(−3)n
for all n ≥ 0. So taking n = 0, 1, 2 gives

a + b + c = x0 = 1
a + 2b − 3c = x1 = 2
a + 4b + 9c = x2 = 1
15 8 3
The solution is a = 20 , b= 20 , c = − 20 , so
1 n+3
xn = 20 (15 + 2 + (−3)n+1 ) n≥0

2. b. The associated polynomial is

p(x) = x3 − 3x + 2 = (x − 1)2 (x + 2)

As 1 is a double root of p(x), [1n ) = [1) and [n1n ) = [n) are solutions to the recurrence by
Theorem 7.5.3. Similarly, [(−2)n ) is a solution, so {[1), [n), [(−2)n)} is a basis for the space
of solutions by Theorem 7.5.4. The required sequence has the form

[xn ) = a[1) + b[n) + c[(−2)n)

for constants a, b, c. Thus, xn = a + bn + c(−2)n for n ≥ 0, so taking n = 0, 1, 2, we get

a + c = x0 = 1
a + b − 2c = x1 = −1
a + 2b + 4c = x2 = 1

The solution is a = 95 , b = − 69 , c = 49 , so
1
5 − 6n + (−2)n+2
 
xn = 9 n≥0

d. The associated polynomial is

p(x) = x3 − 3x2 + 3x − 1 = (x − 1)3

116 Linear Transformations

Hence, [1n) = [1), [n1n) = [n) and [n2 1n ) = [n2 ) are solutions and so [1), [n), [n2 ) is a basis


for the space of solutions. Thus

xn = a · 1 + bn + cn2
a, b, c constants. As x0 = 1, x1 = −1, x2 = 1, we obtain

a = x0 = 1
a + b + c = x1 = −1
a + 2b + 4c = x2 = 1

The solution is a = 1, b = −4, c = 2, so

xn = 1 − 4n + 2n2 n≥0

This can be written

xn = 2(n − 1)2 − 1

3. b. The associated polynomial is

p(x) = x2 − (a + b)x + ab = (x − a)(x − b)

Hence, as a 6= b, {[an), [bn )} is a basis for the space of solutions.

4. b. The recurrence xn+4 = −xn+2 + 2xn+3 has r0 = 0 as there is no term xn . If we write yn = xn+2 ,
the recurrence becomes
yn+2 = −yn + 2yn+1
Now the associated polynomial is x2 − 2x + 1 = (x − 1)2 so basis sequences for the solution
space for yn are [1n ) = [1, 1, 1, 1, . . . ) and [n1n ) = [0, 1, 2, 3, . . . ). As yn = xn+2 , corre-
sponding basis sequences for xn are [0, 0, 1, 1, 1, 1, . . . ) and [0, 0, 0, 1, 2, 3, . . . ). Also,
[1, 0, 0, 0, 0, 0, . . . ) and [0, 1, 0, 0, 0, 0, . . . ) are solutions for xn , so these four sequences
form a basis for the solution space for xn .

7. The sequence has length 2 and associated polynomial x2 + 1. The roots are nonreal: λ1 = i and
λ2 = −i. Hence, by Remark 2,

[in + (−i)n ) = [2, 0, −2, 0, 2, 0, −2, 0, . . . ) and [i(in − (−i)n )) = [0, −2, 0, 2, 0, −2, 0, 2, . . . )

are solutions. They are independent as is easily verified, so they are a basis for the space of solutions.
8. Orthogonality

8.1 Orthogonal Complements and Projections

1. b. Write x1 = (2, 1) and x2 = (1, 2). The Gram-Schmidt algorithm gives

e1 = x1 = (2, 1)
e2 = x2 − x2 ·e12 e1
ke1 k
= (1, 2) − 45 (2, 1)
= 15 {(5, 10) − (8, 4)}
= 35 (−1, 2)

In hand calculations, {(2, 1), (−1, 2)} may be a more convenient orthogonal basis.
d. If x1 = (0, 1, 1), x2 = (1, 1, 1), x3 = (1, −2, 2) then
e1 = x1 = (0, 1, 1)
e2 = x2 − x2 ·e12 e1 = (1, 1, 1) − 22 (0, 1, 1) = (1, 0, 0)
ke1 k
x3 ·e1
e 3 = x3 − e1 − x3 ·e22 e2 = (1, −2, 2) − 02 (0, 1, 1) − 11 (1, 0, 0) = (0, −2, 2)
ke1 k2 ke2 k

2. b. Write e1 = (3, −1, 2) and e2 = (2, 0, −3). Then {e1 , e2 } is orthogonal and so is an orthogonal
basis of U = span {e1 , e2 }. Now x = (2, 1, 6) so take
x·e1
x1 = projU x = e1 + x·e22 e2
ke1 k2 ke2 k
17 14
= 14 (3, −1, 2) − 13 (2, 0, −3)
1
= 182 (271, −221, 1030)
1
Then x2 = x − x1 = 182 (93, 402, 62). As a check: x2 is orthogonal to both e1 and e2 (and so
is in U ⊥ ).
d. If e1 = (1, 1, 1, 1), e2 = (1, 1, −1, −1), e3 = (1, −1, 1, −1) and x = (2, 0, 1, 6), then
{e1 , e2 , e3 } is orthogonal so take

x1 = projU x = x·e1
e1 + x·e22 e2 + x·e32 e3
ke1 k2 ke2 k ke3 k
= 94 (1, 1, 1, 1) − 45 (1, 1, −1, −1) − 34 (1, −1, 1, −1)
1
= 4 (1, 7, 11, 17)

Then, x2 = x − x1 = 41 (7, −7, −7, 7) = 74 (1, −1, −1, 1). Check: x2 is orthogonal to each ei ,
hence x2 is in U ⊥.

117
118 Orthogonality

f. If e1 = (1, −1, 2, 0) and e2 = (−1, 1, 1, 1) then (as x = (a, b, c, d))

x1 = projU x = a−b+2c
6 (1, −1, 2, 0) + −a+b+c+d
4 (−1, 1, 1, 1)
= ( 5a−5b+c−3d
12 , −5a+5b−c+3d
12 , a−b+11c+3d
12 , −3a+3b+3c+3d
12 )
7a+5b−c+3d 5a+7b+c−3d −a+b+c−3d 3a−3b−3c+9d
x2 = x − x1 = ( 12 , 12 , 12 , 12 )

3. a. Write e1 = (2, 1, 3, −4) and e2 = (1, 2, 0, 1), so {e1 , e2 } is orthogonal.

As x = (1, −2, 1, 6)
x·e1
projU x = e + x·e22 e2
ke1 k2 1 ke2 k
= − 30 (2, 1, 3, −4) + 36 (1,
21
2, 0, 1) = 3
10 (−3, 1, −7, 11)

c. projU x = − 15 3
14 (1, 0, 2, −3) + 70 (4, 7, 1, 2) =
3
10 (−3, 1, −7, 11).
4. b. U = span {(1, −1, 0), (−1, 0, 1)} but this basis is not orthogonal. By Gram-Schmidt:
e1 = (1, −1, 0)
e2 = (−1, 0, 1) − (−1, 0, 1)·(1, −1,
2
0)
(1, −1, 0) = − 12 (1, 1, −2)
k(1, −1, 0)k

So we use U = span {(1, −1, 0), (1, 1, −2)}. Then the vector x1 in U closest to x = (2, 1, 0)
is
x1 = projU x = 2−1+0
2 (1, −1, 0) +
2+1+0
6 (1, 1, −2) = (1, 0, −1)
d. The given basis of U is not orthogonal. The Gram-Schmidt algorithm gives
e1 = (1, −1, 0, 1)
e2 = (1, 1, 0, 0) = (1, 1, 0, 0) − 03 e1 = (1, 1, 0, 0)
e3 = (1, 1, 0, 1) − 31 (1, −1, 0, 1) − 22 (1, 1, 0, 0) = 13 (−1, 1, 0, 2)
Given x = (2, 0, 3, 1), we get (using e′3 = (−1, 1, 0, 2) for convenience)
projU x = 33 (1, −1, 0, 1) + 22 (1, 1, 0, 0) + 06 (−1, 1, 0, 2) = (2, 0, 0, 1).
h i h i h i
1 −1 2 1 1 −1 2 1 1 0 −1 1
5. b. Here A = 1 0 −1 1 → 0 1 −3 0 → 0 1 −3 0 . Hence, AxT = 0 has x =
(s − t, 3s, s, t) = s(1, 3, 1, 0) + t(−1, 0, 0, 1).
Thus U ⊥ = span {(1, 3, 1, 0), (−1, 0, 0, 1)}.
8. If x = projU x then x is in U by Theorem 8.1.3. Conversely, if x is in U , let {f1 , . . . , fm } be an
orthogonal basis of U . Then the expansion theorem (applied to the space U ) gives x = ∑i x·fi2 fi =
kfi k
projU x by the definition of the projection.
10. Let {f1 , . . . , fm } be an orthonormal basis of U . If x is in U then, since kfi k = 1 for each i, so
x = (x · f1 )f1 + · · · + (x · fm )fm = projU x by the expansion theorem (applied to the space U ).
 yT 
1
..
14. If {y1 , . . . , ym } is a basis of U ⊥, take A =  . . Then Ax = 0 if and only if yTi x = 0 for each
yTm
0
i; if and only if yi · x = 0 for each i; if and only if x is in (U ⊥ )⊥ = U ⊥⊥ = U . This shows that
U = {x in Rn | Ax = 0}.
 8.2. Orthogonal Diagonalization 119

17. d. If AAT is invertible and E = AT (AAT )−1 A, then

E 2 = AT (AAT )−1 A · AT (AAT )−1 A = AT I(AAT )−1 A = E
T T
E T = AT (AAT )−1 A = AT (AAT )−1 (AT )T
 
−1 −1
= AT (AAT )T A = AT (AT )T AT
 
A
−1
= AT AAT
 
A=E
Thus, E 2 = E = E T .

8.2 Orthogonal Diagonalization

   
3
− 54 1 3 −4
1. b. Since 32 + 42 = 52 , each row has length 5. So 5
4 3 = 5 4 3
is orthogonal.
5 5
√ h i
d. Each row has length a2 + b2 6= 0, so √ 21 2 −ba ba is orthogonal.
a +b
√ √ √
f. The rows have length 6, 3, 2 respectively, so
" √2 √1 √1
# " #
6 6
− 6 2 1 −1
√ √ √
√1 − √13 √1 = √1 2 − 2 2
3 3
6 √ √
0 √1 √1 0 3 3
2 2

is orthogonal.
√ √
h. Each row has length 4 + 36 + 9 = 49 = 7. Hence
" 2 6 3
# " #
7 7 −7 2 6 −3
1
3
7
2
7
6
7 = 7
3 2 6
− 67 3
7
2
7 −6 3 2

is orthogonal.
2. Let P be orthogonal, so P−1 = PT . If P is upper triangular, so also is P−1 , so P−1 = PT is both upper
triangular (P−1 ) and lower triangular PT ). Hence, P−1 = PT is diagonal, whence P = (P−1 )−1 is
diagonal. In particular, P is symmetric so P−1 = PT = P. Thus P2 = I. Since P is diagonal, this
implies that all diagonal entries are ±1.

5. b. cA (x) = x −1 1 x −1 1 = x(x − 2).

Hence the i hare λ1 =i0, λ2 = 2.
h eigenvalues nh io
−1 1 1 −1 1
λ1 = 0: → ; E 0 (A) = span .
h 1 −1 i h 0 i0 nh 1
io
−1
λ2 = 2: 11 11 → 10 10 ; E2 (A) = span 1 .
Note that these eigenvectors are orthogonal (as Theorem 8.2.4 asserts). Normalizing them
gives an orthogonal matrix
 
√1 − √12
h i
1 1 −1
P = √1 2
√1
= √
2 1 1
2 2
h i
0 0
Then P−1 = PT and PT AP = 0 2 .
120 Orthogonality

x−3 0 −7
d. cA (x) = 0 x−5 0 = (x − 5)(x2 − 6x − 40) = (x − 5)(x + 4)(x − 10). Hence the eigen-
−7 0 x−3

 λ1 = 5, λ2= 10,
values are  λ3 = −4.
  
2 −7 0 1 0 0 0
λ1 = 5: 0 0 0 → 0 0 1 ; E5 (A) = span .
1
−7 2 0
  0 0 0
 0
 
7 0 −7 1 0 −1 1
λ2 = 10: 0 5 0 → 0 1 0 ; E10 (A) = span 0 .
 −7 0 7  0 0 0
 1 
−7 0 −7 1 0 1 1
λ3 = −4: 0 −9 0 → 0 1 0 ; E−4 (A) = span 0 .
−7 0 −7 0 0 0 −1
Note that the three eigenvectors are pairwise orthogonal (as Theorem 8.2.4 asserts). Normal-
izing them gives an orthogonal matrix
" #
0 √12 √1
 
2
1 √0 1 1
P = 1 01 0
1
= √2 2 0 0
0 √ − √2 0 1 −1
2

 
5 0 0
Then P−1 = PT and PT AP = 0 10 0 .
0 0 −4
x−5 2 4 x−9 0 9−x x−9 0 0
f. cA (x) = 2 x−8 2 = 2 x−8 2 = 2 x−8 4
4 2 x−5 4 2 x−5 4 2 x−1

= (x − 9) x −2 8 x −4 1
= (x − 9)(x2 − 9x) = x(x − 9)2 .
The eigenvalues
 are λ1 =0, λ2 = 9.       
−5 2 4 1 −4 1 1 −4 1 1 0 −1 2
λ1 = 0 : 2 −8 2 → 0 −18 9 → 0 1 − 21 → ; E0 (A) = span . 0 1 − 21 1
2 −5
 4
  0 1 18 −9 0 0 0
  0 0 0 2
4 2 4 1 2 1 −1 −1
λ2 = 9: 2 1 2 → 0 0 0 ; E9 (A) = span 0 . , 2
4 2 4 0 0 0 1 0
 
−1
However, these are not orthogonal and the Gram-Schmidt algorithm replaces 2 with Z2 =
0
" 2 √−1 √1 # " √ #
2 2 −3 1
 
1 3 2 3 2 √
1
−4 . Hence P = 1
3 0 −4
√
3 2 = √
√
2 0 −4 is orthogonal and satisfies PT AP =
1 2 √1 1 3 2
3
√ 2 2 3 1
  2 3 2
0 0 0
0 9 0 .
0 0 9
   
−2 1
We note in passing that and 2 2 are another orthogonal basis of E9 (A), so Q =
  1
−2 
2 −2 1 0 0 0
1
3
1 2 2 also satisfies QT AQ = 0 9 0 .
2 1 −2 0 0 9

h. To evaluate cA (x), we begin adding rows 2, 3 and 4 to row 1.

x−3 −5 1 −1 x−8 x−8 x−8 x−8
−5 x−3 −1 1 −5 x−3 −1 1
cA (x) = 1 −1 x−3 −5 = 1 −1 x−3 −5
−1 1 −5 x−3 −1 1 −5 x−3
x−8 0 0 0
x+2 4 6
−5 x−2 4 6
= 1 −2 x−4 −6 = (x − 8) −2 x−4 −6
2 −4 x−2
−1 2 −4 x−2
x+2 4 6 x−2 4 6
= (x − 8) x x 0 = (x − 8) 0 x 0
2 −4 x − 2 6 −4 x−2

= x(x − 8) x −6 2 x −6 2 = x(x − 8)(x2 − 4x − 32) = x(x + 4)(x − 8)2

 8.2. Orthogonal Diagonalization 121

−3 −5 1 −1 −3 −5 1 −1 1 −1 −3 −5
" # " # " #
−5 −3 −1 1 −8 −8 0 0 0 −8 −8 −16
λ1 = 0 : 1 −1 −3 −5 → 1 −1 −3 −5 → 0 −16 −24 −40
−1 1 −5 −3 0 0 −8 −8 0 0 1 1
1 0 −2 −3 1 0 0 −1 1
" # " # (" #)
0 1 1 2 0 1 0 1 −1
→ 0 0 1 1 → 0 0 1 1 ; E0 (A) = span −1
0 0 1 1 0 0 0 0 1
−5 −7
1 −1 1 −1 −7 −5 1 0 −3 −2
" # " # " #
−7 −1 −5 1 0 −12 −48 −36
λ2 = −4 : −1 −7 −5
1 → 0 −12 −36 −24 → 00 10 −14 −13
−1
" 1 10 −5 0
−7
1
# 0 0 −12 −12
(" −1 #) 0 0 1 1
→ 00 01 10 −11 ; E−4 (A) = span −1
1

0 0 0 0
" 5 −5 1 −1
# " 1 −1 5 −5
# 1 " 1 −1 0 0 #
−5 5 −1
λ3 = 8 : 1 −1 5 −5
1
→ 00 00 −24 24 −24
24
→ 00 00 10 −10 ;
−1 1 −5 5
(" 1
# " 00 #)0 0 0 0 0 0 0

1
E8 (A) = span 0 , 01
0 1
 1 1 √1
 √
2 − 2 0 
1 −1 2 0
 " 0 0 0 0
2
#
1 1 1 √
 − √ 0  −1 1 2 0 −4
Hence, P =  − 21 −21 02 √1  = 12  −1 −1 0 √2  gives PT AP = 00 0
0
8
0
0 .
2 2 2 √
1 1 √1 1 1 0 2 0 0 0 8
0
2 2 2

x −a 0
x −c −a 0
6. cA (x) = −a x −c =x −c x +a −c x = x(x2 − c2 ) − a2 x = x(x2 − k2 ).
0 −c x

Hence cA (x) = x(x − k)(x + k), where k2 = a2 + c2 , so the eigenvalues are λ1 = 0, λ2 = k, λ3 = −k.
They areall distinct (k6= 0, and
 a 6= 0 or c 6= 0) so the eigenspaces
  are all one dimensional.
0 −a 0 c 0 c
λ1 = 0 : −a 0 −c 0 = 0 ; E0 (A) = span 0
0 −c
 0
  −a  0  −a
k −a 0 a 0 a
λ2 = k : −a k −c k = 0 ; Ek (A) = span k
0 −c k c 0 c
      
−k −a 0 a 0 a
λ3 = −k : −a −k −c −k = 0 ; E−k (A) = span −k
0 −c −k c 0 c
√ √
These eigenvalues
 √ are orthogonal
 and have length, k, 2k, 2k respectively. Hence,
c 2 a a 0 0 0
P= √1 0√ k −k is orthogonal and PT AP = 0 k 0 .
2k −a 2 c c 0 0 −k
h i
10. Similar to Example 8.2.6, q has matrix A = 12 −22 with eigenvalues λ1 = −3 and λ2 = 2 and
h i h i h i
−1 2 1 −1 2
corresponding eigenvectors x1 = and x2 = 1 respectively. Hence P = √ is
h i 2 5 2 1

orthogonal and PT AP = −30 02 . Let

h i h i
y1 −x1 + 2x2
= y = P T
x = √1 ; so y1 = √1 (−x1 + 2x2 ) and y2 = √1 (2x1 + x2 ).
y2 2x1 + x2 5 5 5

Then q = −3y21 + 2y22 is diagonalized by these variables.

11. (c) ⇒ (a). By Theorem 8.2.1 let P−1 AP = D = diag (λ1 , . . . , λn ) where the λi are the eigenvalues
of A. By (c) we have λi = ±1 for each i. It follows that
D2 = diag (λ12 , . . . , λn2 ) = diag (1, . . . , 1) = I
122 Orthogonality

Since A = PDP−1 , we obtain A2 = (PDP−1 )2 = PD2 P−1 = PIP−1 = I. Since A is symmetric, this
proves (a).

13. b. Let A and B be orthogonally similar, say B = PT AP where PT = P−1 . Then B2 = PT APPT AP =
PT AIAP = PT A2 P. Hence A2 and B2 are orthogonally similar.

15. Assume that (Ax) · y = x · Ay for all columns x and y; we must show that AT = A. We have (Ax) · y =
xT AT y and x · Ay = xT Ay, so the given condition asserts that

xT AT y = xT Ay for all columns x and y. (∗)

 
But if E j denotes column j of the identity matrix, then writing A = ai j we have

eTi Ae j = ai j for all i and j.

Since (∗) shows that AT and A have the same (i, j)-entry for each i and j. In other words, AT = A.
Note that the same argument shows that if A and B are matrices with the property that xT By = xT Ay
for all columns x and y, then B = A.
h i h i
cos θ sin θ cos θ sin θ
18. b. If P = − sin θ cos θ and Q = sin θ − cos θ then P and Q are orthogonal matrices, det P = 1
and det Q = −1. (We note that every 2 × 2 orthogonal matrix has the form of P or Q for some
θ .)
d. Since P is orthogonal, PT = P−1 . Hence

PT (I − P) = PT − PT P = PT − I = −(I − PT ) = −(I − P)T

Since P is n × n, taking determinants gives

det PT det (I − P) = (−1)n det [(I − P)T ] = (−1)n det (I − P)

Hence, if I − P is invertible, then det (I − P) 6= 0 so this gives det PT = (−1)n ; that is det P =
(−1)n , contrary to assumption.

21. By the definition of matrix multiplication, the (i, j)-entry of AAT is ri · r j . This is zero if i 6= j,
and equals kri k2 if i = j. Hence, AAT = D = diag (kr1 k2 , kr2 k2 , . . . , krnk2 ). Since D is invertible
(kri k2 6= 0 for each i), it follows that A is invertible and, since row i of AT is a1i a2i · · · a ji · · · ani
 
 . . .
 kr11k2 0 ··· 0
.. .. ..
−1 T −1
··· ···  0 kr21k2 ··· 0 
A =A D =  a 1i ··· a ji ··· a ni   .
 .. .
.. .. .
..

. .. .. . . 
. ··· . ··· . 0 0 ··· 1
krn k2

a ji
Thus, the (i, j)-entry of A−1 is 2 .
kr j k

23. b. Observe first that I − A and I + A commute, whence I − A and (I + A)−1 commute. Moreover,
T  −1
(I + A)−1 = (I + A)T = (I T + AT )−1 = (I − A)−1 . Hence,


PPT = (I − A)(I + A)−1 [(I − A)(I + A)−1 ]T

 8.3. Positive Definite Matrices 123

= (I − A)(I + A)−1 [(I + A)−1 ]T (I − A)T

= (I − A)(I + A)−1 (I − A)−1 (I + A)
= (I + A)−1 (I − A)(I − A)−1 (I + A)
= (I + A)−1 I(I + A)
=I

8.3 Positive Definite Matrices

h i h i h √2 − √12
i √ h i
2 −1 2 −1 T 2 2 −1
1. b. −1 1 → 0 1 . Then A = U U where U = 0 √1
= 2 0 1 .
2 2
   20 4 5
  20 4 5 
20 4 5
d. 4 2 3 → 0 6
5 2 → 0 65 25 .
5 3 5 0 2 154 0 0 12
√ √ √ #
 
√2 √1 5
√
"
5 5 2 5 60 5 12 5 15 5
1 √ √
Hence, U =  0 √6
30
√10
30
=
30
0 6 30 10 30
√
, and A = U T U .
0 0 √5 0 0 5 15
2 15

2. b. If λ k is positive and k is odd, then λ is positive.

4. Assume x 6= 0 is a column. If A and B are positive definite then xT Ax > 0 and xT Bx > 0 so
xT (A + B)x = xT Ax + xT Bx > 0 + 0 = 0
Thus A + B is positive definite. Now suppose r > 0. Then xT (rA)x = r(xT Ax) > 0, proving that rA
is positive definite.
6. Given x in Rn , xT (U T AU )x = (U x)T A(U x) > 0 provided U x 6= 0 (because A is positive definite).
T
Write U = c1 · · · cm where c j in Rn is column j of U . If 0 6= x = x1 · · · xm , then
  

U x = ∑ x j c j 6= 0 because the c j are independent [rank of U is m].

10. Since A is symmetric, the principal axis theorem asserts that an orthogonal matrix P exists such √ that
T
P AP = D = diag (λ1 , λ2 , . . . , λn ) where λi are the √ of A. Since each λi > 0, λi
√ the √ eigenvalues
λ1 , λ2 , . . . , λn . Then B2 = D. As A = PDPT , take


is real and positive, so define B = diag
C = PBPT . Then
C2 = PBPT PBPT = PB2 PT = PDPT = A
Finally, C is symmetric because B is symmetric CT = PT T BT PT = PBPT = C and C has eigen-


√
values λi > 0 (C is similar to B). Hence C is positive definite.
12. b. Suppose that A is positive definite so A = U0T U0 where U0 is upper triangular with positive
diagonal entries d1 , d2 , . . . , dn . Put D0 = diag (d1 , d2 , . . . , dn ). Then L = U0T D−1
0 is lower
−1
triangular with 1’s on the diagonal, U = D0 U0 is upper triangular with 1’s on the diagonal,
and A = LD20U . Take D = D20 .
Conversely, if A = LDU as in (a), then AT = U T DLT . Hence, AT = A implies that U T DLT =
LDU , so U T√= L √ and LT = √ U by
(a). Hence, A = U T DU . If D = diag (d1 , d2 , . . . , dn ), let
D1 = diag d1 , d2 , . . . , dn . Then D = D21 so A = U T D21U = (D1U )T (D1U ). Hence, A
is positive definite.
124 Orthogonality

8.4 QR-Factorization
h i h i
2 1
1. b. The columns of A are c1 = 1and c2 = 1 . First apply the Gram-Schmidt algorithm
h i
f1 = c1 = 21
h i h i h i
c2 ·f1 1 2 − 51
f2 = c2 − f = − 53 =
kf1 k2 1 1 1 2
5

Now normalize to obtain

h i
1 2
q1 = = √1
kf1 k f1 5 1
h i
1 −1
q2 = = √1
kf2 k f2 5 2
h i
2 −1
√1
 
Hence Q = q1 q2 = 1 5 2 is an orthogonal matrix. We obtain R from equation
(8.5) preceding Theorem 8.4.1:
√
√3
h i h 5
i h i
L = kf01 k c2 · q1
kf2 k = 0 √1
5
= √1
5
5
0
3
1
5

Then A = QR.
 T  T
d. The columns of A are c1 = 1 −1 0 1 , c2 = 1 0 1 −1 and
 T
c3 = 0 1 1 0 . Apply the Gram-Schmidt algorithm
 T
f1 = c1 = 1 −1 0 1
T 0 T
f2 = c2 − c2 ·f12 f1 = 1 0 1 −1
 
− 3 F1 = 1 0 1 −1
kf1 k
c3 ·f1
f3 = c3 − f1 − c3 ·f22 f2
kf1 k2 kf2 k
T −1  T T
− 31
 
= 0 1 1 0 − 3 1 −1 0 1 1 0 1 −1
T
= 23 0 1 1 1


Normalize
1
T
√1

Q1 = kf1 k f1 = 1 −1 0 1
3
1 1
 T
Q2 = kf2 k f2 = √ 1 0 1 −1
3
1 1
 T
Q3 = kf3 k f3 = √
3
0 1 1 1
1 1 0
" #
−1 0 1
√1
 
Hence Q = q1 q2 q3 = 3 0 1 1 has orthonormal columns. We obtain R from
1 −1 1
equation (8.5) preceding Theorem 8.4.1:
√ −1
 " √1
# " #

kf1 k c2 · q1 c3 · q1 3 √
3 0 −1
√3 3
R= 0 kf2 k c3 · q2 = 0 3 √1
3 = √1 0 3 1
0 0 kf3 k √2
3
0 0 0 0 2
3

Then A = QR.
 8.5. Computing Eigenvalues 125

2. b. If A = QR is a QR-factorization of A, then R has independent columns (it is invertible) as does

Q (its columns are orthonormal). Hence A has independent columns by (a). The converse is
by Theorem 8.4.1.

8.5 Computing Eigenvalues

h i
1. b. A = 5
−3
2
−2. Then cA (x) = x −3 5 x−2
+ 2 = (x + 1)(x − 4), so λ1 = −1, λ2 = 4.
h i h i h i
If λ1 = −1: −63 −21 → 30 10 ; eigenvector = −13 .
h i h i h i
−1 −2 1 2 2
If λ2 = 4: 3 6 → ; dominant eigenvector = −1 .
h i 0 0
Starting with x0 = 11 , the power method gives x1 = Ax0 , x2 = Ax1 , . . . :
h i h i h i h i
7 25 103 409
x1 = −5 , x2 = −11 , x3 = −53 , x4 = −203

h i
2
These are approaching (scalar multiples of) the dominant eigenvector −1 . The Rayleigh
x ·x
quotients are rk = k k+1 , k = 0, 1, 2, . . . , so r0 = 1, r1 = 3.29, r2 = 4.23, r3 = 3.94. These
kxk k2
are approaching the dominant eigenvalue 4.
h i
− 3 −1
√
d. A = 31 10 ; cA (x) = x−1 x = x2 − 3x − 1, so the eigenvalues are λ1 = 21 (3 + 13),
√ √
λ2 = 12 (3 − 13). Thus the dominant eigenvalue is λ1 = 21 (3 + 13). Since λ1 λ2 = −1 and
λ1 + λ2 = 3, we get h i h i
λ1 − 3 −1 1 −λ1
−1 λ1 → 0 0
h i h i
so a dominant eigenvector is λ11 . We start with x0 = 11 .
Then xk+1 = Axk , k = 0, 1, . . . gives
h i h i h i h i
x1 = 41 , x2 = 134 , x3 = 43 13 , x 4 = 142
43

h i h i
λ1 3.302776
These are approaching scalar multiples of the dominant eigenvector 1 = 1 . The
xk ·xk+1
Rayleigh quotients are rk = :
kxk k2

r0 = 2.5, r1 = 3.29, r2 = 3.30270, r3 = 3.30278

These are rapidly approaching the dominant eigenvalue λ1 = 3.302776.

h i
− 3 −1
√ 
b. A = 31 10 ; cA (x) = x−1 = x2 − 3x − 3; λ1 = 12 3 + 13 = 3.302776 and

2. x
√ 
λ2 = 12 3 − 13 = −0.302776. The QR-algorithm proceeds as follows:

h i h i h i
A1 = 31 10 = Q1 R1 where Q1 = √110 31 −31 , R1 = √110 100 13 .
126 Orthogonality

h i h i h i
1 33 1 1 33 1 109 3
A 2 = R 1 Q1 = = Q2 R2 where Q2 = √1090 √1
10 1 −3 1 −33 , R2 = 1090 0 10 .
h i h i
360 1
1
A3 = R2 Q2 = 109 1 −33 = 3.302752 0.009174
0.009174 −0.302752 .
The diagonal entries already approximate λ1 and λ2 to 4 decimal places.

4. We prove that ATk = Ak for each k by induction in k. If k = 1, then A1 = A is symmetric by hypothesis,

so assume ATk = Ak for some k ≥ 1. We have Ak = Qk Rk so Rk = Q−1 T
k Ak = Qk Ak because Qk is
orthogonal. Hence
Ak+1 = Rk Qk = QTk Ak Qk
so
ATk+1 = (QTk Ak Qk )T = QTk ATk QTk T = QTk Ak Qk = Ak+1
The eigenvalues of A are all real as A is symmetric, so the QR-algorithm asserts that the Ak converge
to an upper triangular matrix T . But T is symmetric (it is the limit of symmetric matrices), so it is
diagonal.

8.6 Singular Value Decomposition

4. b. t σ1 , . . . , t σr .

7. If A = U ΣV T then Σ is invertible, so A−1 = V Σ−1U T is a SVD.

8. b. First AT A = In so ΣA = In .
   
√1
1 1
1 0 √1 1 1
A = 2 1 −1
0 1 2 −1 1
   
1 −1 √1 −1 1
= √12
1 1 2 1 1
 
−1 0
=
0 1

9. b.
A=F
1 1 1 1
" #
h ih i
1 3 4 20 0 0 0 1 1 −1 1 −1
= 5 4 −3 0 10 0 0 2 1 1 −1 −1
1 −1 1 −1

13. b. If x ∈ Rn then xT (G + H)x = xT Gx + xT Hx ≥ 0 + 0 = 0.

 1 1

4 0 4
17. b.
− 14 0 − 14
 8.7. Complex Matrices 127

8.7 Complex Matrices

q √
|1 − i|2 + |1 + i|2 + 12 + (−1)2 =
p
1. b. (1 + 1) + (1 + 1) + 1 + 1 = 6
q √
d. 4 + |−i|2 + |1 + i|2 + |1 − i|2 + |2i|2 = 4 + 1 + (1 + 1) + (1 + 1) + 4 = 13
p

2. b. Not orthogonal: h(i, −i, 2 + i), (i, i, 2 − i)i = i(−i) + (−i)(−i) + (2 + i)(2 + i) = 3 + 4i
d. Orthogonal: h(4 + 4i, 2 + i, 2i), (−1 + i, 2, 3 − 2i)i = (4 + 4i)(−1 − i) + (2 + i)2 + (2i)(3 +
2i) = (−8i) + (4 + 2i) + (−4 + 6i) = 0.

3. b. Not a subspace. For example, i(0, 0, 1) = (0, 0, i) is not in U .

d. If v = (v + w, v − 2w, v) and w = (v′ + w′ , v′ − 2w′ , v′ ) are in U then

v + w = ((v + v′ ) + (w + w′ ), (v + v′ ) − 2(w + w′ ), (v + v′ )) is in U
zv = (zv + zw, zv − 2zw, zv) is in U
0 = (0 + 0, 0 − 20, 0) is in U

Hence U is a subspace.

4. b. Here U = {(iv + w, 0, 2v − w) | v, w ∈ C} = {v(i, 0, 2) + w(1, 0, −1) | v, w ∈ C} =

span {(i, 0, 2), (1, 0, −1)}.
If z(i, 0, 2) + t(1, 0, −1) = (0, 0, 0) with z, t ∈ C, then iz + t = 0, 2z − t = 0. Adding gives
(2 + i)z = 0, so z = 0; and so t = −iz = 0. Thus {(i, 0, 2), (1, 0, −1)} is independent over C,
and so is a basis of U . Hence dim C U = 2.
d. U = {(u, v, w) | 2u + (1 + i)v − iw = 0; u, v, w ∈ C)}. The condition is w = −2iu + (1 − i)v,
so
U = {(u, v, −2iu + (1 − i)v) | u, v ∈ C} = span {(1, 0, −2i), (0, 1, 1 − i)}
If z(1, 0, −2i) + t(0, 1, i − 1) = (0, 0, 0) then components 1 and 2 give z = 0 and t = 0. Thus
{(1, 0, −2i), (0, 1, 1 − i)} is independent over C, and so is a basis of U . Hence dim C U = 2.
h i h i h i
5. b. A = −32 32 , AH = AT = 23 −32 , A−1 = 13 1 2 −3
3
H
2 . Hence, A is not hermitian (A 6= A )
and not unitary (A−1 6= AH ). However, AAH = 13I = AH A, so A is normal.
h i h iT h i
−i −i
d. A = 1i −1 , AH = (A)T = −i1 −1i = 1i −1 = A. Thus A is hermitian and so is
normal. But, AAH = A2 = 2I so A is not unitary.
h i h i
1 1+i T H 1 1−i
f. A = 1 + i i . Here A = A so A = A = 1 − i −i 6= A (thus A is not hermitian). Next,
h i h i
AAH = 2 +3 i 2 −3 2i 6= I so A is not unitary. Finally, AH A = 2 −3 2i 2 +3 2i 6= AAH , so A is not
normal. h i h i h i
h. A = √1 z z
. Here A = √1 z z
so AH = √1 z z
. Thus A = AH if and only
2|z| z −z 2|z| z −z 2|z| z −z
h 2 i
if z = z; that is A is hermitian if and only if z is real. We have AAH = 1 2 2 |z|
0
0
2 |z| 2 = I,
2|z|
and similarly, AH A = I. Thus it is unitary (and hence normal).
128 Orthogonality

h i h i
4 3−i x−4 −3 + i
8. b. A = 3+i , cA (x) = −3 = x2 − 5x − 6 = (x + 1)(x − 6).
1
h −i x−1 i h i h i
−5 −3 + i 3+i 2 −2
Eigenvectors for λ1 = −1: −3 − i → 0 ; an eigenvector is x1 = .
h 1
i h 0
i h 3+i i
Eigenvectors for λ2 = 6: −32− i −35+ i → 02 −30+ i ; an eigenvector is x2 = 3 −2 i .
√ 1
h
−2 3−i
i
As x1 and x2 are orthogonal and kx1 k = kx2 k = 14, U = √ is unitary and
h i 14 3 + i 2

U H AU = −10 06 .
h i
d. A = 1 −2 i 1 +3 i ; cA (x) = −1 x−2
+i
−1 − i
x−3 = x2 − 5x + 4 = (x − 1)(x − 4).
h i h i h i
Eigenvectors for λ1 = 1: −1−1+ i −1−2− i → 10 1 +0 i ; an eigenvector is x1 = 1−1 +i
.
h i h i h i
Eigenvectors for λ2 = 4: −12+ i −11− i → −10+ i 01 ; an eigenvector is x2 = 1 −1 i .
√ h
+i
i
Since x1 and x2 are orthogonal and kx1 k = kx2 k = 3, U = √13 1−1 1
1−i is unitary and
h i
U H AU = 10 04 .
 
1 0 0 x−1 0 0
f. A = 0 1 1 + i ; cA (x) = 0 x−1 −1 − i = (x − 1)(x2 − 3x) = (x − 1)x(x − 3).
0 1−i −1 + i
2
 0
 x − 2   
0 0 0 0 1 0 1
Eigenvectors for λ1 = 1: 0 0 −1 − i → 0 0 1 ; an eigenvector is x1 = 0 .
0 −1 + i
   −1  0 0 0   0
−1 0 0 1 0 0 0
If λ2 = 0: 0 −1 −1 − i → 0 1 1 + i ; an eigenvector is x2 = 1 + i .
0 −1 + i −i 0 0 −1
 0    
2 0 0 1 0 0 0
Eigenvectors for λ3 = 3: 0 2 −1 − i → 0 −1 + i 1 ; an eigenvector is x3 = 1 .
0 −1 + i 1 0 0 0 1−i
 √
√

3 0 0
Since {x1 , x2 , x3 } is orthogonal and kx2 k = kx3 k = 3, U = √1 0 1+i 1 is orthog-
3 0 −1 1−i
 
1 0 0
onal and U ∗AU = 0 0 0 .
0 0 3

10. b. (1) If z = (z1 , z2 , . . . , zn ) then kzk2 = |z1 |2 + |z2 |2 + · · · + |zn |2 . Thus kzk = 0 if and only if
|z1 | = · · · = |zn | = 0, if and only if z = (0, 0, . . . , 0).
(2) By Theorem 8.7.1, we have hλ z, wi = λ hz, wi and hz, λ wi = λ hz, wi. Hence

kλ zk2 = hλ z, λ zi = λ hz, λ zi = λ λ hz, zi = |λ |2 kzk2

Taking positive square roots gives kλ zk = |λ | kzk.

b. If A is hermitian then A = AT . If A = ai j , the (k, k)-entry of A is akk , and the (k, k)-entry of
 
11.
AT is akk . Thus, A = AT implies that akk = akk for each k; that is akk is real.

14. b. Let B be skew-hermitian, that is BH = −B. Then Theorem 8.7.3 gives

(B2 )H = (BH )2 = (−B)2 = B2 , so B2 is hermitian

(iB)H = (−i)BH = (−i)(−B) = iB, so iB is hermitian.

d. If Z = A + B where AH = A and BH = −B, then Z H = AH + BH = V − B. Solving gives

Z + Z H = 2V and Z − Z H = 2B, so V = 12 (Z + Z H ) and S = 12 (Z + Z H ). Hence the matrices A
 8.8. An Application to Linear Codes over Finite Fields 129

and B are uniquely determined by the conditions Z = A + B, AH = A, BH = −B, provided such

A and B exist. But always,
Z = 12 (Z + Z H ) + 12 (Z − Z H )
and the matrices A = 12 (Z + Z H ) and B = 12 (Z − Z H ) are hermitian and skew-hermitian respec-
tively:

AH = 12 (Z H + Z HH ) = 21 (Z H + Z) = A
BH = 12 (Z H − Z HH ) = 21 (Z H − Z) = −B

16. b. If U is unitary, then U −1 = U H . We must show that U −1 is unitary, that is (U −1 )−1 = (U −1)H .
But
(U −1)−1 = U = (U H )H = (U −1 )H
h i h i h i
1 i 1 −i T i −1
18. b. If V = −i 0 then V is hermitian because V = i 0 = V , but iV = 1 0 is not
hermitian (it has a nonreal entry on the main diagonal).
h i h i
0 1 a b
21. b. Given A = −1 0 , let U = c d be invertible and real, and assume that U −1AU =
h i h i
λ µ λ µ
0 ν . Thus, AU = U 0 ν so
h i h i
c d aλ aµ + bν
−a −b = cλ cµ + d ν

Equating first column entries gives c = aλ and −a = cλ . Thus, −a = (aλ )λ = aλ 2 so (1 +

λ 2 )a = 0. Now λ is real (a and c are not both zero so either λ = ac or λ = − ac ), so 1 + λ 2 6= 0.
Thus a = 0 (because (1 + λ 2 )a = 0) whence c = aλ = 0. This contradicts the assumption that
A is invertible.

8.8 An Application to Linear Codes over Finite Fields

1. b. The elements with inverses are 1, 3, 7, 9: 1 and 9 are self-inverse; 3 and 7 are inverses of each
other. As for the rest, 2 · 5 = 4 · 5 = 6 · 5 = 8 · 5 = 0 in Z10 so 2, 5, 4, 6 and 8 do not have
inverses in Z10 .
d. The powers of 2 computed in Z10 are: 2, 4, 8, 16 = 6, 32 = 2, . . . , so the sequence repeats: 2,
4, 8, 16, 2, 4, 8, 16, . . . .

2. b. If 2a = 0 in Z10 then 2a = 10k for some integer k. Thus a = 5k so a = 0 or a = 5 in Z10 .

Conversely, it is clear that 2a = 0 in Z10 if a = 0 or a = 5.

3. b. We want a number a in Z19 such that 11a = 1. We could try all 19 elements in Z19 , the one
that works is a = 7. However the euclidean algorithm is a systematic method for finding a. As
in Example 8.8.2, first divide 19 by 11 to get

19 = 1 · 11 + 8
130 Orthogonality

Then divide 11 by 8 to get

11 = 1 · 8 + 3
Now divide 8 by 3 to get
8 = 2·3+2
Finally divide 3 by 2 to get
3 = 1·2+1
The process stops here since a remainder of 1 has been reached. Now eliminate remainders
from the bottom up:

1 = 3 − 1 · 2 = 3 − (8 − 2 · 3) = 3 · 3 − 8
= 3(11 − 1 · 8) − 8 = 3 · 11 − 4 · 8
= 3 · 11 − 4(19 − 1 · 11) = 7 · 11 − 4 · 19

Hence 1 = 7 · 11 − 4 · 19 = 7 · 11 in Z19 because 19 = 0 in Z19 .

b. Workinghin Z7 , weihavehdet A = −1 −1
6. i 15h− 24 =i 1 + 4 = 5 6= 0, so A exists. Since 5 = 3 in Z7 ,
A−1 = 3 −43 −65 = 3 33 15 = 22 31 .

7. b. Gaussian elimination works over any field F in the same way that we have been using it over
R. In this case we have F = Z7 , and we reduce the augmented matrix of the system as follows.
We have 5 · 3 = 1 in Z7 so the first step in the reduction is to multiply row 1 by 5 in Z7 :
h i h i h i h i h i
3 1 4 3 1 5 6 1 1 5 6 1 1 5 6 1 1 0 5 3
4 3 1 1 → 4 3 1 1 → 0 4 5 4 → 0 1 3 1 → 0 1 3 1

Hence x and y are the leading variables, and the non-leading variable z is assigned as a param-
eter, say z = t. Then, exactly as in the real case, we obtain x = 3 + 2t, y = 1 + 4t, z = t where t
is arbitrary in Z7 .

9. b. If the inverse is a + bt then 1 = (1 + t)(a + bt) = (a − b) + (a + b)t. This certainly holds if

a − b = 1 and a + b = 0. Adding gives 2a = 1, that is −a = 1 in Z3 , that is a = −1 = 2. Hence
a + b = 0 gives b = −a = 1, so a + bt = 2 + t; that is (1 + t)−1 = 2 + t. Of course it is easily
checked directly that (1 + t)(2 + t) = 1.

10. b. The minimum weight of C is 5, so it detects 4 errors and corrects 2 errors by Theorem 8.8.5.

11. b. The linear (5, 2)-code {00000, 01110, 10011, 11101} has minimum weight 3 so it corrects 1
error by Theorem 8.8.5.

12. b. The code is {0000000000, 1001111000, 0101100110, 0011010111,

1100011110, 1010101111, 0110110001, 1111001001}.
This has minimum distance 5 and so corrects 2 errors.

13. b. C = {00000, 10110, 01101, 11011} is a linear (5, 2)-code of minimal weight 3, so it corrects
single errors.
h i
  u
14. b. G = 1 u where u is any nonzero vector in the code. H = In−1 .
 8.9. An Application to Quadratic Forms 131

8.9 An Application to Quadratic Forms

h 1
i h i
1 2 (1 − 1) 1 0
1. b. A = 1 = 0 2
2 (−1 + 1) 2
" 1 1
# " #
1 2 (2 + 4) 2 (−1 + 5) 1 3 2
d. A = 1
2 (4 + 2) 1 1
2 (0 − 2) = 3 1 −1
1 1
2 (5 − 1) 2 (−2 + 0) 3 2 −1 3
h i
−1 −2
2. b. q = xT Ax where A = 12 21 . cA (x) = x−2 2
x − 1 = x − 2x − 3 = (x + 1)(x − 3)
h i h i h i
2 −2 1 −1
λ1 = 3: −2 2 → 0 0 ; so an eigenvector is x1 = 11 .
h i h i h i
λ2 = −1: −2 −2
−2 −2 → 1 1
; so an eigenvector is x 2 = 1
.
h i 0 0 h i−1
Hence, P = √1 11 −11 is orthogonal and PT AP = 30 −10 .
2 h ih i h i
T 1 1 1 x1 1 x1 + x2
As in Theorem 8.9.1, take y = P x = 2 1 −1
√
x2 = 2 x1 − x2 . Then
√

y1 = √1 (x1 + x2 ) and y2 = √1 (x1 − x2 )

2 2

Finally, q = 3y21 − y22 , the index of q is 1 (the number of positive eigenvalues) and the rank of q
is 2 (the number of nonzero eigenvalues).
 
7 4 4
T
d. q = x Ax where A = 4 1 −8 . To find cA (x), subtract row 2 from row 3:
4 −8 1

x−7 −4 −4 x−7 −4 −4
cA (x) = −4 x−1 8 = −4 x−1 8
−4 8 x−1 0 −x + 9 x−9
x−7 −8 −4
= −4 x+7 8 = (x − 9)2 (x + 9)
0 0 x−9
       
2 −4 −4 1 −2 −2
2 2
λ1 = 9: −4 →8 8 ; orthogonal eigenvectors
0 0 2 ,
0 −1 .
−4 8 8 0 0 −1
0
      2 
−16
−4 −4 4 1 1 4 0 2 −1
λ2 = −9: −48 → 0 −9 9 → 0 1 −1 ; eigenvector
−10 2 . These
−4
−10 8 0 9 −9 0 0 0
 2
2 2 −1
eigenvectors are orthogonal and each has length 3. Hence, P = 31 2 −1 2 is orthog-
−1 2 2
 
9 0 0
onal and PT AP = 0 9 0 . Thus
0 0 −9
    
2 2 −1 x1 2x1 + 2x2 − x3
T 1 1
y=P x= 3
2 −1 2 x2 = 3
2x1 − x2 + 2x3
−1 2 2 x3 −x1 + 2x2 + 2x3

so
y1 = 31 [2x1 + 2x2 − x3 ]
y2 = 13 [2x1 − x2 + 2x3 ]
y3 = 13 [−x1 + 2x2 + 2x3 ]
will give q = 9y21 + 9y22 − 9y23 . The index of q is 2 and the rank of q is 3.
132 Orthogonality

 
5 −2 −4
f. q = xT Ax where A = −2 8 −2 . To find cA (x), subtract row 3 from row 1:
−4 −2 5

x−5 2 4 x−9 0 −x + 9
cA (x) = 2 x−8 2 = 2 x−8 2
4 2 x−5 4 2 x−5
x−9 0 0
= 2 x−8 4 = x(x − 9)2
4 2 x−1

       
4 2 4 2 1 2 −2 1
λ1 = 9: →
2 1 2 ; orthogonal eigenvectors are
0 0 0 2 and 2 .
4 2 4 0 0 0 1 −2
       
−5 2
4 1 −4 1 1 0 −1 2
λ2 = 0: 2 −8
2 → 0 −18 9 → 0 2 −1 ; an eigenvector is 1 .
4 −5
2 0 18 −9 0 0 0 2
 
−2 1 2
These eigenvectors are orthogonal and each has length 3. Hence P = 13 2 2 1 is or-
1 −2 2
 
9 0 0
T
thogonal and P AP = 0 9 0 . If
0 0 0

  
−2 2 1 x1
T 1
y=P x= 3
1 2 −2 x2
2 1 2 x3

then

y1 = 13 (−2x1 + 2x2 + x3 )
y2 = 13 (x1 + 2x2 − 2x3 )
y3 = 13 (2x1 + x2 + 2x3 )

gives q = 9y21 + 9y22 . The rank and index of q are both 2.

 
1 −1 0
T
h. q = x Ax where A = −1 0 1 . To find cA (x), add row 3 to row 1:
0 1 1

x−1 1 0 x−1 0 x−1

cA (x) = 1 x −1 = 1 x −1
0 −1 x−1 0 −1 x−1
x−1 0 0
= 1 x −2 = (x − 1)(x − 2)(x + 1)
0 −1 x−1

       
1 1 0 1 1 0 1 0 1 −1
λ1 = 2: 1 2 −1 → 0 1 −1 → 0 1; an eigenvector is
−1 . 1
0 −1 1 0 −1 1 0 0 0 1
       
0 1 0 1 1 −1 1 0 −1 1
λ2 = 1: 1 1 −1 → 0 1 0 → 0 1 0 ; an eigenvector is 0 .
0 −1 0 0 0 0 0 0 0 1
       
−2 1 0 1 −1 −1 1 0 1 1
λ3 = −1: 1 −1 −1 → 0 −1 −2 → 0 1 2 ; an eigenvector is 2 .
0 −1 −2 0 −1 −2 0 0 0 −1
Hence,
√ √
 
− √13 √1 √1
" #
2 6 − 2 3 1
√
P= √1
3
0 √2 = √1 2 0 2
6 √ √
6
√1 √1 − √16 2 3 −1
3 2
 8.9. An Application to Quadratic Forms 133

 
2 0 0
is orthogonal and PT AP = 0 1 0 . If
0 0 −1

 √ √ √  
−√ 2 2 √2 x1
y=P x= T √1 3 0 3 x2
6 1 2 −1 x3

then

y1 = √1 (−x1 + x2 + x3 )
3
1
y2 = √ (x1 + x3 )
2
y3 = √1 (x1 + 2x2 − x3 )
6

gives q = 2y21 + y22 − y23 . Here q has index 2 and rank 3.

h i h i
3. b. q = 3x2 − 4xy = xT Ax where x = xy , A = −23 −20 . cA (t) = t −2 3 2t = (t − 4)(t + 1)
h i h i h i
λ1 = 4: 12 24 → 10 20 ; an eigenvector is −12 .
h i h i h i
−4 2 2 −1
λ2 = −1: 2 −1 → 0 0 ; an eigenvector is 12 .
h i h i h i
Hence, P = √1 −12 12 gives PT AP = 40 −10 . If y = PT x = xy11 , then x1 = √1 (2x − y)
5 5
and y1 = √1 (x + 2y).
The equation q = 2 becomes 4x21 − y21 = 2, a hyperbola.
5
h i h i
2 2 T x 2 2
d. q = 2x + 4xy + 5y = x Ax where x = y , A = 2 5 .
−2 −2
In this case cA (t) = t −2 t − 5 = (t − 1)(t − 6).
h i h i h i
λ1 = 6: −24 −21 → 20 −10 ; an eigenvector is 12 .
h i h i h i
λ2 = 1: −1 −2 −4
−2
→ 1 2
0 0 ; an eigenvector is 2
−1 .
h i h i h i
1 1 2 6 0 x1
T
Hence, P = 5 2 −1 gives P AP = 0 1 . If y = P x =
√ T , then x1 = √1 (x + 2y),
y1 5
y1 = √1 (2x − y) and q = 1 becomes 6x21 + y21 = 1. This is an ellipse.
5
h i h i
4. After the rotation, the new variables x1 = xy11 are related to x = xy by x = Ax1 where A =
h i
cos θ − sin θ
sin θ cos θ (this is equation (8.8) preceding Theorem 8.9.2, or see Theorem 2.6.4). Thus x =
x1 cos θ −y1 sin θ and y = x1 sin θ +y1 cos θ . If these are substituted in the equation ax2 +bxy+cy2 =
d, the coefficient of x1 y1 is

−2a sin θ cos θ + b(cos2 θ − sin2 θ ) + 2c sin θ cos θ = b cos 2θ − (a − c) sin 2θ .

This is zero if θ is chosen so that

a−c b
cos 2θ = √ and sin 2θ = √
b2 +(a−c)2 b2 +(a−c)2

 2  2
Such an angle 2θ exists because √ 2a−c 2 + √ 2b 2 = 1.
b +(a−c) b (a−c)
134 Orthogonality

   
x1 1 2 −2
xT Ax + Bx =
 
7. b. The equation is 7 where x = x2 ,A= 2 3 0 , B = 5 0 −6 .
x3 −2 0 3

t −1 −2 2 t −1 −2 2 t −1 −4 2
cA (x) = −2 t −3 0 = −2 t −3 0 = −2 t −3 0
2 0 t −3 0 t −3 t −3 0 0 t −3

= (t − 3)(t 2 − 4t − 5) = (t − 3)(t − 5)(t + 1)

     
2 −2 2 1 0 0 0
λ1 = 3: −2 0 0 → 0 1 −1 ; an eigenvector is 1 .
2 0 0
   0 0 0
     1
4 −2 2 −2 2 0 1 0 1 1
λ2 = 5: −2 2 0 → 0 2 2 → 0 1 1 ; an eigenvector is 1 .
−1
2 0 2  0 2 2
 0 0 0   
−2 −2 2 1 1 −1 1 0 −2 2
λ3 = −1: −2 −4 0 → 0 −2 −2 → 0 1 1 ; an eigenvector is −1 .
2 0 −4 0 −2 −2 0 0 0 1
√
 
√1 √2
" #
0 3 6 0 2 2

3 0 0

√ √
Hence, P =  √1
2
− √13 − √16  = √1 3 2 −1 T
satisfies P AP = 0 5 0 . If
6 √ √ 0 0 −1
√1 + √13 √1 3 − 2 1
2 6

   √ 
y1 √ 3(x2 + x3 )
y= y2 T
=P x= √1 2(x1 + x2 − x3 )
y3 6 2x1 − x2 + x3

then

y1 = √1 (x2 + x3 )
2
1
y2 = √ (x1 + x2 − x3 )
3
y3 = √1 (2x1 − x2 + x3 )
6

As P−1 = PT , we have x = Py so substitution in xT Ax + Bx = 7 gives

yT (PT AP)y + (BP)y = 7

√ √  h √ √ √ i
As BP = √1 −6 3 11 2 4 = −3 2 113 3 2 3 6 , this is

6
√ √ √
3y21 + 5y22 − y23 − (3 2)y1 + ( 11
3 3)y2 + ( 2
3 6)y3 = 7

9. b. We have A = U T U where U is upper triangular with positive diagonal entries. Hence

q(x) = xT U T U x = (U x)T (U x) = kU xk2

So take y = U x as the new column of variables.

 8.10. An Application to Constrained Optimization 135

8.10 An Application to Constrained Optimization

This section contains no exercises.

8.11 An Application to Statistical Principal Component

Analysis

This section contains no exercises.

9. Change of Basis

9.1 The Matrix of a Linear Transformation

 
a
1. b. CB (v) = 2b − c because v = ax2 + bx + c = ax2 + (2b − c)(x + 1) + (c − b)(x + 2).
c−b
 
a−b
1
d. CB (v) = 2
a+b because
−a + 3b + 2c

v = (a, b, c) = 12 [(a − b)(1, −1, 2) + (a + b)(1, 1, −1) + (−a + 3b + 2c)(0, 0, 1)]

h i
2 1 3
CD [T (1)] CD [T (x)] CD [T (x2 )]
 
2. b. MDB (T ) = = −1 0 −2 . Comparing columns gives
h i h i h i
2 1 3
CD [T (1)] = −1 CD [T (x)] = 0 CD [T (x2 )] = −2

Hence

T (1) = 2(1, 1) − (0, 1) = (2, 1)

T (x) = 1(1, 1) + 0(0, 1) = (1, 1)
T (x2 ) = 3(1, 1) − 2(0, 1) = (3, 1)

Thus

T (a + bx + cx2 ) = aT (1) + bT (x) + cT (x2 )

= a(2, 1) + b(1, 1) + c(3, 1)
= (2a + b + 3c, a + b + c)
n h h io n h io n h io n h io i
1 0 0 1 0 0
3. b. MDB (T ) = CD T 0 0 CD T 0 0 CD T 1 0 CD T 00 0
1
h nh io nh io nh io nh io i
1 0 0 0 0 1 0 0
= CD CD CD CD
" 1 0 00 0
0
# 1 0 0 0 0 1

0 0 1 0
= 0 1 0 0
0 0 0 1

d. MDB (T ) =  CD [T (1)] CD [T (x)] CD [T (x2 )]

 

=  CD (1) CD (x + 1) CD (x2 + 2x + 1)


1 1 1
= 0 1 2
0 0 1

137
138 Change of Basis

1 2
" #
    5 3
4. b. MDB (T ) = CD [T (1, 1)] CD [T (1, 0)] CD (1, 5, 4, 1) CD (2, 3, 0, 1)
= = 4 0 .
1 1
h i
b
We have v = (a, b) = b(1, 1) + (a − b)(1, 0) so CB (v) = a − b . Hence,

1 2 2a − b
" # " #
h i
5 3 b 3a + 2b
CD [T (v)] = MDB (T )CB (v) = 4 0 a−b = 4b
1 1 a

Finally, we recover the action of T :

T (v) = (2a − b)(1, 0, 0, 0) + (3a + 2b)(0, 1, 0, 0) + 4b(0, 0, 1, 0) + a(0, 0, 0, 1)

= (2a − b, 3a + 2b, 4b, a)

d. MDB (T ) =  CD [T (1)] CD [T (x)] CD [T (x2 )]

 

= h CD (1, 0) iCD (1, 0) CD (0, 1)

1 1
− 12
= 2
1
2
1 1
h2 2 2 i
= 12 11 11 −11
 
a
We have v = a + bx + cx2 so CB (v) = b . Hence
c

h i a
 h i
1 1 1 −1 1 a+b−c
CD [T (v)] = MDB (T )CB (v) = 2 1 1 1
b = 2 a+b+c
c

Finally, we recover the action of T :

T (v) = 12 (a + b − c)(1, −1) + 12 (a + b + c)(1, 1) = (a + b, c).

n hh io n h io n h io n h io i
f. MDB (T ) = CD T 10 00 CD T 00 10 CD T 01 0
0 CD T 0
0
0
1
h h i h i h i h i i
1 0 0 1 0 1
= CD 0 0 CD 1 0 CD 1 0 CD 00 0
1
" 1 0 0 0 #
0 1 1 0
= 0 1 1 0 .
0 0 0 1
h i h i h i h i h i
a b 1 0 0 1 0 0 0 0
We have v = c d =a 0 0 +b 0 0 +c 1 0 +d 0 1 ,
1 0 0 0
" a
# " #" a
# " a
#
b 0 1 1 0 b b+c
so CB (v) = c . Hence CD [T (v)] = MDB (T )CB (v) = 0 1 1 0 c = b+c .
d 0 0 0 1 d d
Finally, we recover the action of T :
h i h i h i h i h i
b+c
T (v) = a 0 0 + (b + c) 00
1 0 1
0 + (b + c) 0
1
0
0 +d 0
0
0
1 = a
b+c d

T S
5. b. Have R3 → R4 → R2 . Let B, D, E be the standard bases. Then
 
MED (S) = CE [S(1, 0, 0, 0)] CE [S(0, 1, 0, 0)] CE [S(0, 0, 1, 0)] CE [S(0, 0, 0, 1)]
 
= CE (1, 0) CE (1, 0) CE (0, 1) CE (0, −1)
 9.1. The Matrix of a Linear Transformation 139

h i
1 1 0 0
= 0 0 1 −1
 
MDB (T ) = CD [T (1, 0, 0)] CD [T (0, 1, 0)] CD [T (0, 0, 1)]
 
= CD (1, 0, 1, −1) CD (1, 1, 0, 1) CD (0, 1, 1, 0)
" 1 1 0 #
0 1 1
= 1 0 1
−1 1 0

We have ST (a, b, c) = S(a + b, c + b, a + c, b − a) = (a + 2b + c, 2a − b + c). Hence

 
MEB (ST ) = CE [ST (1, 0, 0)] CE [ST (0, 1, 0)] CE [ST (0, 0, 1)]
 
= CE (1, 2) CE (2, −1) CE (1, 1)
h i
= 12 −12 11

With this we confirm Theorem 9.1.3 as follows:

" 1 1 0
#
h i h i
MED (S)MDB(T ) = 10 10 01 −10 0
1
1
0
1
1 = 1
2
2
−1
1
1 = MEB (ST )
−1 1 0

T S
d. Have R3 → P2 → R2 with bases B = {(1, 0, 0), (0, 1, 0), (0, 0, 1)}, D = 1, x, x2 ,


E = {(1, 0), (0, 1)}.

MED (S) = CE [S(1)] CE [S(x)] CE [S(x2 )]

 
 
= CE (1, 0) CE (−1, 0) CE (0, 1)
h i
= 10 −10 10

 
MDB (T ) = CD [T (1, 0, 0)] CD [T (0, 1, 0)] CD [T (0, 0, 1)]
= CD (1 − x) CD (−1 + x2 ) CD (x)
 
 
1 −1 0
= −1 0 1
0 1 0

The action of ST is ST (a, b, c) = S (a − b) + (c − a)x + bx2 = (2a − b − c, b). Hence,

 

 
MEB (ST ) = CE [ST (1, 0, 0)] CE [ST (0, 1, 0)] CE [ST (0, 0, 1)]
 
= CE (2, 0) CE (−1, 1) CE (−1, 0)
h i
2 −1 −1
= 0 1 0

Hence, we verify Theorem 9.1.3 as follows:

h i 1 −1 0
 h i
1 −1 0 2 −1 −1
MED (S)MDB(T ) = 0 0 1 −1 0 1 = 0 1 0 = MEB (ST )
0 1 0
140 Change of Basis

 
7. b. MDB (T ) =  CD [T (1, 0, 0)] CD [T (0, 1, 0)] CD [T (0,
 0, 1)]
=  CD (0, 1, 1) CD (1, 0, 1) CD (1, 1, 0)
0 1 1
= 1 0 1
1 1 0
If T −1 (a, b, c) = (x, y, z) then (a, b, c) = T (x, y, z) = (y + z, x + z, x + y). Hence, y + z = a,
x + z = b, x + y = c. The solution is
T −1 (a, b, c) = (x, y, z) = 12 (b + c − a, a + c − b, a + b − c)
Hence,
MBD (T −1 ) = CB [T −1 (1, 0, 0)] CB [T −1 (0, 1, 0)] CB [T −1 (0, 0, 1)]
 

= CB − 12 , 12 , 21 CB 12 , − 12 , 12 CB 12 , 21 , − 12




 
−1 1 1
1
=2 1 −1 1
1 1 −1

This matrix is MDB (T )−1 as Theorem 9.1.4 asserts.

d. MDB (T ) =  CD [T (1)] CD [T (x)] C[T (x2 )]
 

=  CD (1, 0, 0) CD (1, 1, 0) CD (1, 1, 1)
1 1 1
= 0 1 1
0 0 1
If T −1 (a,b, c) = r + sx + tx2 , then (a, b, c) = T (r + sx + tx2 ) = (r + s + t, s + t, t). Hence,
r + s + t = a, s + t = b, t = c; the solution is t = c, s = b − c, r = a − b. Thus,
T −1 (a, b, c) = r + sx + tx2 = (a − b) + (b − c)x + cx2
Hence,
MBD (T −1 ) = CB [T −1 (1, 0, 0)] CB [T −1 (0, 1, 0)] CB [T −1 (0, 0, 1)]
 

= CB (1) CB (−1 + x) CB (−x + x2 )

 
 
1 −1 0
= 0 1 −1
0 0 1

This matrix is MDB (T )−1 as Theorem 9.1.4 asserts.

h n h io n h io n h io n h io i
8. b. MDB (T ) = CD T 10 00 CD T 00 10 CD T 01 00 CD T 00 0
1
 
= " CD (1, 0, 0,#0) CD (1, 1, 0, 0) CD (1, 1, 1, 0) CD (0, 0, 0, 1)
1 1 1 0
0 1 1 0
= 0 0 1 0
0 0 0 1
This is invertible and the matrix inversion algorithm (and Theorem 9.1.4) gives
" 1 −1 0 0 #
MDB (T −1 ) = [MDB (T )]−1 = 0
0
1
0
−1
1
0
0
0 0 0 1

If v = (a, b, c, d) then
1 −1 0 0 a−b
" #" a
# " #
0 1 −1 0 b−c
CB [T −1 (v)] = MDB (T −1 )CD (v) = 0 0 1 0
b
c = c
0 0 0 1 d d
 9.1. The Matrix of a Linear Transformation 141

Hence, we get a formula for the action of T −1 :

h i h i h i h i
−1 −1 1 0 0 1 0 0 0 0
T (a, b, c, d) = T (v) = (a − b) 0 0 + (b − c) 0 0 +c 1 0 +d 0 1
h i
= a −c b b −d c

 
12. Since D = {T (e1 ), . . . , T (en )}, we have CD T (e j ) = C j = column j of In . Hence,
 
MDB (T ) = CD [T (e1 )] CD [T (e2 )] · · · CD [T (en )]
 
= C1 C2 · · · Cn = In

16. b. Define T : Pn → Rn+1 by T [p(x)] = (p(a0 ), p(a1 ), . . . , p(an )), where a0 , . . . , an are fixed
distinct real numbers. If B = {1, x, . . . , xn } and D ⊆ Rn+1 is the standard basis,

MDB (T ) = [ CD [T (1)] CD [T (x)] CD T (x2 ) · · · CD [T (xn )] ]

 

= [ CD (1, 1, . . . , 1) CD (a0 , a1 , . . . , an ) CD (a20 , a21 , . . . , a2n ) · · · CD (an0 , an1 , . . . , ann ) ]

2
 n

1 a0 a0 ... a0
1 a1 a21 ... an1
=
 
.. .. .. .. .. 
. . . . .
1 an a2n ... ann

Since the ai are distinct, this matrix has nonzero determinant by Theorem 3.2.7. Hence, T is
an isomorphism by Theorem 9.1.4.
R S, T
20. d. Assume that V → W → U . Recall that the sum S + T : W → U of two operators is defined by
(S + T ) (w) = S(w) + T (w) for all w in W . Hence, for v in V :

[(S + T ) R] (v) = (S + T )[R(v)]

= S[R(v)] + T [R(v)]
= (SR)(v) + (T R)(v)
= (SR + T R)(v)

Since this holds for all v in V , it shows that (S + T )R = SR + T R.

21. b. If P and Q are subspaces of a vector space W , recall that P + Q = {p + q | p in P, q in Q} is a

subspace of W (Exercise 25 Section 6.4). Now let w be any vector in im (S + T ). Then w =
(S + T )(v) = S(v) + T (v) for some v in V , whence w is in im S + im T . Thus, im (S + T ) ⊆
im S + im T .

22. b. If T is in X10, then T (v) = 0 for all v in X1 . As X ⊆ X1 , this implies that T (v) = 0 for all v in
X ; that is T is in X 0. Hence, X10 ⊆ X 0 .

24. b. We have R : V → L(R, V ) defined by R(v) = Sv . Here Sv : R → V is defined by Sv (r) = rv.

R is a linear transformation: The requirements that R(v + w) = R(v) + R(w) and R(av) =
aR(v) translate to Sv+w = Sv + Sw and Sav = aSv . If r is arbitrary in R:

Sv+w (r) = r(v + w) = rv + rw = Sv (r) + Sw (r) = (Sv + Sw )(r)

142 Change of Basis

Sav (r) = r(av) = a(rv) = a [Sv (r)] = (aSv)(r)

Hence, Sv+w = Sv + Sw and Sav = aSv so R is linear.

R is one-to-one: If R(v) = 0 then Sv = 0 is the zero transformation R → V . Hence we have
0 = Sv (r) = rv for all r; taking r = 1 gives v = 0. Thus ker R = 0.
R is onto: Given T in L(R, V ), we must find v in V such that T = R(v); that is T = Sv . Now
T : R → V is a linear transformation and we take v = T (1). Then, for r in R:

Sv (r) = rv = rT (1) = T (r · 1) = T (r)

Hence, Sv = T as required.

25. b. Given the linear transformation T : R → V and an ordered basis B = {b1 , b2 , . . . , bn } of

V , write T (1) = a1 b1 + a2 b2 + · · · + an bn where the ai are in R. We must show that T =
a1 S1 + a2 S2 + · · · + an Sn where Si (r) = rbi for all r in R. We have

(a1 S1 + a2 S2 + · · · + an Sn )(r) = a1 S1 (r) + a2S2 (r) + · · · + an Sn (r)

= a1 (rb1 ) + a2 (rb2 ) + · · · + an (rbn )
= rT (1)
= T (r)

for all r in R. Hence a1 S1 + a2 S2 + · · · + an Sn = T .

27. b. Given v in V , write v = r1 b1 + r2 b2 + · · · + rn bn , ri in R. We must show that r j = E j (v) for

each j. To see this, apply the linear transformation E j :

E j (v) = E j (r1 b1 + r2 b2 + · · · + r j b j + · · · + rn bn )
= r1 E j (b1 ) + r2 E j (b2 ) + · · · + r j E j (b j ) + · · · + rn E j (bn )
= r1 · 0 + r2 · 0 + · · · + r j · 1 + · · · + rn · 0
= rj

using the definition of E j .

9.2 Operators and Similarity

− 23 1
   
−1 2
−3 −2 1
1
(x2 )
 
1. b. PD←B = CD (x) CD (1 + x) CD = 1 1 0 = 2
2 2 0 because
0 0 1 0 0 2

x = − 32 · 2 + 1(x + 3) + 0(x2 − 1)
1 + x = (−1) · 2 + 1(x + 3) + 0(x2 − 1)
x2 = 12 · 2 + 0(x + 3) + 1(x2 − 1)
 9.2. Operators and Similarity 143

Given v = 1 + x + x2 , we have
− 12
   
0
CB (v) = 1 and CD (v) = 1
1 1

because v = 0 · x + 1(1 + x) + 1 · x2 and v = − 12 · 2 + 1 · (x + 3) + 1(x2 − 1). Hence

    
−3 −2 1 0 −1
1 1
PD←BCB (v) = 2 2 2 0 1 =2 2 = CD (v)
0 0 2 1 2

as expected.
 
1 1 −1
(1 + x + x2 ) (−1 + x2 )
 
4. b. PB←D = CB CB (1 − x) CB = 1 −1 0
1 0 1
 
 1 1 1 1
2

PD←B = CD (1) CD (x) CD (x ) = 3 1 −2 1 because
−1 −1 2

1 2 2
 
1= 3 (1 + x + x ) + (1 − x) − (−1 + x )
1 2 2
 
x= 3 (1 + x + x ) − 2(1 − x) − (−1 + x )
x2 = 1 2 2
 
3 (1 + x + x ) + (1 − x) + 2(−1 + x )

The fact that PD←B = (PB←D )−1 is verified by multiplying these matrices. Next:
 
1 0 1
2 2
 
PE←D = CE (1 + x + x ) CE (1 − x) CE (−1 + x ) = 1 −1 0
0 1 −1
 
0 0 1
PE←B = CE (1) CE (x) CE (x2 ) = 0 1 0
 
1 0 0

where we note the order of the vectors in E = x2 , x, 1 . Finally, matrix multiplication verifies


that PE←D PD←B = PE←B .

5. b. Let B = {(1, 2, −1), (2, 3, 0), (1, 0, 2)} be the basis formed by the transposes of the columns
of A. Since D is the standard basis:
 
  1 2 1
PD←B = CD (1, 2, −1) CD (2, 3, 0) CD (1, 0, 2) = 2 3 0 =A
−1 0 2

Hence Theorem 9.2.2 gives

 
6 −4 −3
−1 −1
 
A = (PD←B ) = PB←D = CB (1, 0, 0) CB (0, 1, 0) CB (0, 0, 1) = −4 3 2
3 −2 −1

because

(1, 0, 0) = 6(1, 2, −1) − 4(2, 3, 0) + 3(1, 0, 2)

(0, 1, 0) = −4(1, 2, −1) + 3(2, 3, 0) − 2(1, 0, 2)
(0, 0, 1) = −3(1, 2, −1) + 2(2, 3, 0) − 1(1, 0, 2)
144 Change of Basis

b. Since B0 = 1, x, x2 , we have

7.
 
1 1 0
(1 − x2 ) (2x + x2 )
 
P = PB0 ←B = CB0 CB0 (1 + x) CB0 = 0 1 2
−1 0 1

CB0 [T (1)] CB0 [T (x)] CB0 [T (x2 )]

 
MB0 (T ) =
CB0 (1 + x2 ) CB0 (1 + x) CB0 (x + x2 )
 
=
 
1 1 0
= 0 1 1
1 0 1

Finally

CB [T (1 − x2 )] CB [T (1 + x)] CB [T (2x + x2 )]
 
MB (T ) =
= CB (1 − x) CB (2 + x + x2 ) CB (2 + 3x + x2 )
 
 
−2 −3 −1
= 3 5 3
−2 −2 0

because

1 − x = −2(1 − x2 ) + 3(1 + x) − 2(2x + x2 )

2 + x + x2 = −3(1 − x2 ) + 5(1 + x) − 2(2x + x2 )
2 + 3x + x2 = −1(1 − x2 ) + 3(1 + x) + 0(2x + x2 )

The verification that P−1 MB0 (T )P = MB (T ) is equivalent to checking that MB0 (T )P = PMB (T ),
and so can be seen by matrix multiplication.
h ih ih i h ih i h i
8. b. P−1 AP = −75 −23 29 −12
70 −29
3 2
7 5 = 5 −2
7 −3
3 2
7 5 = 1 0
0 −1 .
nh i h io
3
Let B = 7 , 25 consist of the columns of P. These are eigenvectors of A correspond-
ing to the eigenvalues 1, −1 respectively. Hence,
h  h i  h i i h h i h i i h i
MB (TA ) = CB TA 37 CB TA 25 = CB 37 CB −2 −5 = 1
0 −1
0

9. b. Choose a basis of R2 , say B = {(1, 0), (0, 1)}, and compute

h i
    3 5
MB (T ) = CB [T (1, 0)] CB [T (0, 1)] = CB (3, 2) CB (5, 3) = 2 3

−3 −5
Hence, cT (x) = cMB (T ) (x) = x−2 x−3 = x2 − 6x − 1. Note that the calculation is easy
because B is the standard basis, but any basis could be used.
d. Use the basis B = 1, x, x2 of P2 and compute


MB (T ) = CB [T (1)] CB [T (x)] CB [T (x2 )]

 

= CB (1 + x − 2x2 ) CB (1 − 2x + x2 ) CB (−2 + x)
 
 9.2. Operators and Similarity 145

 
1 1 −2
= 1 −2 1
−2 1 0

Hence,
x−1 −1 2 x−1 −1 2
cT (x) = cMB (T ) (x) = −1 x+2 −1 = −1 x+2 −1
2 −1 x −x + 3 0 x−2

= x3 + x2 − 8x − 3
nh i h i h i h io
1 0 0 1 0 0 0 0
f. Use B = 0 0 , 0 0 , 1 0 , 0 1 and compute
h
n h io n h io n h io n h io i
MB (T ) = CB T 10 00 CB T 00 10 CB T 01 0
0 CB T 0
0
0
1
h h i h i h i h i i
= CB 11 00 CB 00 11 CB −1 0
−1 0 CB 0
0 −1
−1
" 1 0 −1 0 #
0 1 0 −1
= 1 0 −1 0
0 1 0 −1

Hence,
x−1 0 1 0
0 x−1 0 1
cT (x) = cMB (T ) (x) = −1 0 x+1 0
0 −1 0 x+1
x−1 0 1 0 x−1 1
= (x − 1) 0 x+1 0 + −1 0 0 = x4
−1 0 x+1 0 −1 x+1

12. Assume that A and B are both n ×n and that null A = null B. Define TA : Rn → Rn by TA (x) = Ax for
all x in Rn ; similarly for TB . Then let T = TA and S = TB . Then ker S = null B and ker T = null A
so, by Exercise 28 Section 7.3 there is an isomorphism R : Rn → Rn such that T = RS. If B0 is the
standard basis of Rn , we have

A = MB0 (T ) = MB0 (RS) = MB0 (R)MB0 (S) = U B

where U = MB0 (R). This is what we wanted because U is invertible by Theorem 9.1.4.
Conversely, assume that A = U B with U invertible. If x is in null A then Ax = 0, so U Bx = 0,
whence Bx = 0 (because U is invertible), that is x is in null B. In other words null A ⊆ null B. But
B = U −1 A so null B ⊆ null A by the same argument. Hence null A = null B.

16. b. We verify first that S is linear. Showing S(w+v) = S(w)+S(v) means showing that MB (Tw+v ) =
MB (Tw ) + MB (Tv ). If B = {b1 , b2 } then column j of MB (Tw+v ) is

CB [Tw+v (b j )] = CB [(w + v)b j ] = CB (wb j + vb j ) = CB (wb j ) +CB (vb j )

because CB is linear. This is column j of MB (Tw ) + MB (Tv ), which shows that S(w + v) =
S(w) + S(v). A similar argument shows that MB (Taw ) = aMB (Tw ), so S(aw) = aS(w), and
hence that S is linear.
To see that S is one-to-one, let S(w) = 0; by Theorem 7.2.2 we must show that w = 0. We
have MB (Tw ) = S(w) = 0 so, comparing jth columns, we see that CB [Tw (b j )] = CB [wb j ] = 0 for
146 Change of Basis

j = 1, 2. As CB is an isomorphism, this means that wb j = 0 for each j. But B is a basis of C and

1 is in C, so there exist r and s in R such that 1 = rb1 + sb2 . Hence w = w1 = rwb1 + swb2 = 0,
as required.
Finally, to show that S(wv) = S(w)S(v) we first show that Tw Tv = Twv . Indeed, given z in C,
we have
(Tw Tv )(z) = Tw (Tv (z)) = w(vz) = (wv)z = Twv (z)
Since this holds for all z in C, it shows that Tw Tv = Twv . But then Theorem 9.2.1 shows that

S(wv) = MB (Tw Tv ) = MB (Tw )MB (Tv ) = S(w)S(v)

This is what we wanted.

9.3 Invariant Subspaces and Direct Sums

2. b. Let v ∈ T (U ), say v = T (u) where u ∈ U . Then T (v) = T [T (u)] ∈ T (U ) because T (u) ∈ U .

This shows that T (U ) is T -invariant.

3. b. Given v in S(U ), we must show that T (v) is also in S(U ). We have v = S(u) for some u in U .
As ST = T S, we compute:

T (v) = T [S(u)] = (T S)(u) = (ST )(u) = S[T (u)]

As T (u) is in U (because U is T -invariant), this shows that T (v) = S[T (u)] is in S(U ).

6. Suppose that a subspace U of V is T -invariant for every linear operator T : V → V ; we must show
that either U = 0 or U = V . Assume that U 6= 0; we must show that U = V . Choose u 6= 0 in U , and
(by Theorem 6.4.1) extend {u} to a basis {u, e2 , . . . , en } of V . Now let v be any vector in V . Then
(by Theorem 7.1.3) there is a linear transformation T : V → V such that T (u) = v and T (ei ) = 0
for each i. Then v = T (u) lies in U because U is T -invariant. As v was an arbitrary vector in V , it
follows that V = U .
[Remark: The only place we used the hypothesis that V is finite dimensional is in extending {u} to
a basis of V . In fact, this is true for any vector space, even of infinite dimension.]

8. b. We have U = span 1 − 2x2 , x + x2 . To show that U is T -invariant, it suffices (by Example



9.3.3) to show that T (1 − 2x2 ) and T (x + x2 ) both lie in U . We have

T (1 − 2x2 ) = 3 + 3x − 3x2 = 3(1 − 2x2 ) + 3(x + x2 )


(∗)
T (x + x2 ) = −1 + 2x2 = −(1 − 2x2 )

So both T (1 − 2x2 ) and T (x + x2 ), so U is T -invariant. To get a block triangular matrix for T

extend the basis 1 − 2x2 , x + x2 of U to a basis B of V in any way at all, say

B = 1 − 2x2 , x + x2 , x2

 9.3. Invariant Subspaces and Direct Sums 147

Then, using (∗), we have

 
3 −1 1
MB (T ) = CB T (1 − 2x2 ) CB T (x + x2 ) CB T (x2 ) =

      
3 0 1
0 0 3

where the last column is because T (x2 ) = 1 + x + 2x2 = (1 − 2x2 ) + (x + x2 ) + 3(x2 ). Finally,
x−3 1 −1
x−3 1
cT (x) = −3 x −1 = (x − 3) −3 x = (x − 3)(x2 − 3x + 3)
0 0 x−3

9. b. Algebraic Solution. If U is TA -invariant and U 6= {0}, U 6= R2 , then dim U = 1. Thus U = Ru

where u 6= 0. Thus TA (u) is in Ru (because U is T -invariant), say TA (u) = ru, that is Au = ru,
whence (rI − A) u = 0. But
r − cos θ − sin θ
det (rI − A) = sin θ r − cos θ = (r − cos θ )2 + sin2 θ 6= 0 as sin θ 6= 0(0 < θ < π )

Hence, (rI − A)u = 0 implies u = 0, a contradiction. So U = 0 or U = R2 .

Geometric Solution. If we view R2 as the euclidean plane, and U 6= 0, R2 , is a TA -invariant
subspace, then U must have dimension 1 and so be a line through the origin (Example 5.2.13).
But TA is rotation through θ counterclockwise about the origin (Theorem 2.6.4), so it will move
the line U unless θ = 0 or θ = π , contrary to our assumption that 0 < θ < π . So no such line
U can exist.

10. b. If v is in U ∩W , then v = (a, a, b, b) = (c, d, c, −d) for some a, b, c, d. Hence a = c, a = d,

b = c and b = −d. It follows that d = −d so a = b = c = d = 0; that is U ∩W = {0}. To see
that R4 = U +W , we have (after solving systems of equations)

(1, 0, 0, 0) = 12 (1, 1, −1, −1) + 12 (1, −1, 1, 1) is in U +W

(0, 1, 0, 0) = 12 (1, 1, 1, 1) + 12 (−1, 1, −1, −1) is in U +W
(0, 0, 1, 0) = 21 (−1, −1, 1, 1) + 12 (1, 1, 1, −1) is in U +W
(0, 0, 0, 1) = 21 (1, 1, 1, 1) + 12 (−1, −1, −1, 1) is in U +W

Hence, R4 = U +W . A simpler argument is as follows. As dim U = 2 = dim W , the subspace

U ⊕W has dimension 2 + 2 = 4 by Theorem 9.3.6. Hence U ⊕W = R4 because dim R4 = 4.
h i h i
a a c d
d. If A is in U ∩W , then A = b b = −c d for some a, b, c, d, whence a = b = c = d = 0.
Thus, U ∩W = {0}. Thus, by Theorem 9.3.7

dim (U ⊕W ) = dim U + dim W = 2 + 2 = 4

h dimiMh22 = 4,iweh have U

Since i ⊕W
h =M i 22 . Again, as in (b), we could show directly that each
1 0 0 1 0 0 0 0
of 0 0 , 0 0 , 1 0 , 0 1 is in U +W .

14. First U is a subspace because 0E = 0, and AE = A and A1 E = A1 implies that

(A + A1 )E = AE + A1 E = A + A1 and (rA)E = r(AE) = rA for all r ∈ R

148 Change of Basis

Similarly, W is a subspace because 0E = 0, and BE = 0 = B1 E implies that we have (B + B1 )E =

BE + B1 E = 0 + 0 = 0 and (rB)E = r(BE) = r0 = 0 for all r ∈ R.
These calculations hold for any matrix E; but if E 2 = E we get Mnn = U ⊕W . First U ∩W = {0}
because X in U ∩ W implies X = X E because X is in U and X E = 0 because X is in W , so X =
X E = 0. To prove that U +W = Mnn let X be any matrix in Mnn . Then:

X E is in U because (X E)E = X E 2 = X E
X − X E is in W because (X − X E)E = X E − X E 2 = X E − X E = 0.

Hence X = X E + (X − X E) where X E is in U and (X − X E) is in W ; that is X is in U +W . Thus

Mnn = U +W .
17. By Theorem 6.4.5, we have dim (U ∩W ) + dim (U +W ) = dim U + dim W = n by hypothesis. So
if U +W = V then dim (U +W ) = n, whence dim (U ∩W ) = 0. This means that U ∩W = {0} so,
since U +W = V , we have proved that V = U ⊕W .
18. b. First, ker TA is TA -invariant by Exercise 2. Now suppose that U is any TA -invariant subspace,
U 6= 0, U 6= R2 . Then dim U = 1, say U = Rp, p 6= 0. Thus p is in U so Ap = TA (p) is in
U , say Ap = λ p where λ is a real number. Applying A again, we get A2 p = λ Ap = λ 2 p. But
A2 = 0, so this gives 0 = λ 2 p. Thus λ 2 = 0, whence λ = 0 and Ap = λ p = 0. Hence p is in
ker TA , whence U ⊆ ker TA . But dim U = 1 = dim ( ker TA ), so U = ker TA .
20. hLet B1 be ai basis of U and extend it (using Theorem 6.4.1) to a basis B of V . Then MB (T ) =
MB1 (T ) Y
0 Z by Theorem 9.3.1. Since we are writing T1 for the restriction of T to U , MB1 (T ) =
MB1 (T1 ). Hence,
h i
cT (x) = det [xI − MB (T )] = det xI − M0B1 (T ) xI−x
−Z

= det [xI − MB1 (T1 )] det [xI − Z] = cT1 (x) · q(x)

where q(x) = det [xI − Z].

22. b. We have T : P3 → P3 given by T [p(x)] = p(−x) for all p(x) in P3 . We leave it to the reader to
verify that T is a linear operator. We have

T 2 [p(x)] = T {T [p(x)]} = T [p(−x)] = p(−(−x)) = p(x) = 1P3 (p(x))

Hence, T 2 = 1P3 . As in Example 9.3.10, let

U1 = {p(x) | T [p(x)] = p(x)} = {p(x) | p(−x) = p(x)}

U2 = {p(x) | T [p(x)] = −p(x)} = {p(x) | p(−x) = −p(x)}

These are the subspaces of even and odd polynomials in P3 , respectively, and have bases
B1 = {1, x2 } and B2 = {x, x3 }. Hence, use the ordered basis B = {1, x2 , x, x3 } of P3 . Then
h i h i
MB (T ) = MB10(T ) MB 0(T ) = I02 −I0 2
2

as in Example 9.3.10. More explicitly,

MB (T ) = CB [T (1)] CB [T (x2 )] CB [T (x)] CB [T (x3 )]

 
 9.3. Invariant Subspaces and Direct Sums 149

CB (1) CB (x2 ) CB (−x) CB (−x3 )

 
=
" 1 0 0 0 #
0 1 0 0
= 0 0 −1 0
0 0 0 −1
h i
I2 0
= 0 −I2

d. Here T 2 (a, b, c) = [−(−a + 2b + c) + 2((b + c) + (−c), (b + c) − c), −(−c)] = (a, b, c), so

T 2 = 2 R3 .
Note that T (1, 1, 0) = (1, 1, 0), while T (1, 0, 0) = −(1, 0, 0) and T (0, 1, −2) = −(0, 1, −2).
Let B1 = {(1, 1, 0)} and B2 = {(1, 0, 0), (0, −1, 2)}. These are bases of U1 = R(1, 1, 0) and
U2 = R(1, 0, 0)+R(0, 1, −2), respectively. So if we take B = {(1, 1, 0), (1, 0, 0),(0, −1, 2)}
h i h i 1 0 0
−1 0 MB1 (T ) 0
then MB1 (T ) = [1] and MB2 (T ) = 0 −1 . Hence MB (T ) = 0 MB (T ) = 0 −1 0 .
2 0 0 −1

23. b. Given v, T [v − T (v)] = T (v) − T 2 (v) = T (v) − T (v) = 0, so v − T (v) lies in ker T . Hence
v = (v − T (v)) + T (v) is in ker T + im T for all v, that is V = ker T + im T . If v lies in
ker T ∩ im T , write v = T (w), w in V . Then 0 = T (v) = T 2 (w) = T (w) = v, so ker T ∩ im T =
0.
25. b. We first verify that T 2 = T . Given (a, b, c) in R3 , we have

T 2 (a, b, c) = T (a + 2b, 0, 4b + c) = (a + 2b, 0, 4b + c) = T (a, b, c)

Hence T 2 = T . As in the preceding exercise, write

U1 = {v | T (v) = v} and U2 = {v | T (v) = 0} = ker (T ).

Then we claim that R3 = U1 ⊕U2 . To show R3 = U1 +U2 , observe that v = T (v) + [v − T (v)]
for each v in R3 , and T (v) is in U1 [because T [T (v)] = T 2 (v) = T (v)] while v − T (v) is in U2
[because T [v − T (v)] = T (v) − T 2 (v) = 0]. Finally we show that U1 ∩U2 = {0}. For if v is in
U1 ∩U2 then T (v) = v and T (v) = 0 so certainly v = 0.
Next, we show that U1 and U2 are T -invariant. If v is in U1 then T (v) is also in U1 because
T [T (v)] = T 2 (v) = T (v). Similarly U2 is T -invariant because, if v is in U2 , that is T (v) = 0,
then T [T (v)] = T 2 (v) = R(v) = 0; that is T (v) is also in U2 .
It is clear that T (a, b, c) = (a, b, c) if and only if b = 0; that is U1 = {(a, 0, c) | b, c in R},
so B1 = {(1, 0, 0), (0, 0, 1)} is a basis of U1 . Since T (v) = v for all v in U1 the restriction of
T to U1 is the identity transformation on U1 , and so has matrix I2 .
Similarly, T (a, b, c) = (0, 0, 0) holds if and only if a = −2b and c = −4b for some b, so
U1 = R(2, −1, 4) and B2 = {(2, −1, 4)} is a basis of U2 . Clearly the restriction of T to U2 is
the zero transformation, and so has matrix 02 — a 1 × 1 matrix.
Finally then, B = B1 ∪ B2 = {(1, 0, 0), (0, h0, 1), (2, −1, 4)} 3
i ish a basisiof R (since we have
shown that R3 = U1 ⊕U2 ), so T has matrix MB10(T ) MB 0(T ) = I02 001 .
2

29. b. We have T f2, z [v] = T f , z [T f , z (v)] = T f , z [ f (v)z] = f [ f (v)z]z = f (v) f (z)z. This expression
equals T f , z (v) = f (v)z for all v if and only if

f (v)(z − f (z)z) = 0
150 Change of Basis

for all v. Since f 6= 0, f (v) 6= 0 for some v, so this holds if and only if

z = f (z)z

As z 6= 0, this holds if and only if f (z) = 1.

30. b. Let λ be an eigenvalue of T . If A is in Eλ (T ) then T (A) = λ A; that is UA = λ A. If we write

A = p1 p2 · · · pn in terms of its columns p1 , p2 , . . . , pn , then UA = λ A becomes
   
U p1 p2 · · · pn = λ p1 p2 · · · pn
   
U p1 U p2 · · · U pn = λ p1 λ p2 · · · λ pn

Comparing columns gives U pi = λ pi for each i; that is pi is in Eλ (U ) for each i. Conversely,

if p1 , p2 , . . . , pn are all in Eλ (U ) then U pi = λ pi for each i, so T (A) = UA = λ A as above.
Thus A is in Eλ (T ).
10. Inner Product Spaces

10.1 Inner Products and Norms

1. b. P5 fails: h(0, 1, 0), (0, 1, 0)i = −1

The other axioms hold. Write x = (x1 , x2 , x3 ), y = (y1 , y2 , y3 ) and z = (z1 , z2 , z3 ).
P1 holds: hx, yi = x1 y1 − x2 y2 + x3 y3 is real for all x, y in Rn .
P2 holds: hx, yi = x1 y1 − x2 y2 + x3 y3 = y1 x1 − y2 x2 + y3 x3 = hy, xi
P3 holds: hx + y, zi = (x1 + y1 )z1 − (x2 + y2 )z2 + (x3 + y3 )z3
= (x1 z1 − x2 z2 + x3 z3 ) + (y1 z1 − y2 z2 + y3 z3 ) = hx, zi + hy, zi
P4 holds: hrx, yi = (rx1 )y1 − (rx2 )y2 + (rx3 )y3 = r(x1 y1 − x2 y2 + x3 y3 ) = rhx, yi
d. P5 fails: hx − 1, x − 1i = 0 · 0 = 0
P1 holds: hp(x), q(x)i = p(1)q(1) is real.
P2 holds: hp(x), q(x)i = p(1)q(1) = q(1)p(1) = hq(x), p(x)i
P3 holds: hp(x) + r(x), q(x)i = [p(1) + r(1)]q(1) = p(1)q(1) + r(1)q(1)
= hp(x), q(x)i + hr(x), q(x)i
P4 holds: hrp(x), q(x)i = [rp(1)]q(1) = r[p(1)q(1)] = rhp(x), q(x)i
f. P5 fails: Here h f , f i = 2 f (0) f (1) for any f , so if f (x) = x − 21 then h f , f i = − 12 .
P1 holds: h f , gi = f (1)g(0) + f (0)g(1) is real.
P2 holds: h f , gi = f (1)g(0) + f (0)g(1) = g(1) f (0) + g(0) f (1) = hg, f i
P3 holds: h f + h, gi = ( f + h)(1)g(0) + ( f + h)(0)g(1)
= [ f (1) + h(1)]g(0) + [ f (0) + h(0)]g(1)
= [ f (1)g(0) + f (0)g(1)] + [h(1)g(0) + h(0)g(1)] = h f , gi + hh, gi
P4 holds: hr f , hi = (r f )(1)g(0) + (r f )(0)g(1) = [r · f (1)]g(0) + [r f (0)]g(1)
= r[ f (1)g(0) + f (0)g(1)] = rh f , gi

2. If h , i denotes the inner product on V , then hu1 , u2 i is a real number for all u1 and u2 in U .
Moreover, the axioms P1 − P5 hold for the space U because they hold for V and U is a subset of V .
So h , i is an inner product for the vector space U .
π
3. b. k f k2 = −ππ cos2 x dx = −ππ 12 [1 + cos(2x)] dx = 21 x + 12 sin(2x) −π = π . Hence fˆ = √1π f is a
R R 

unit vector.
h i h ih i
d. kvk2 = hv, vi = vT −11 −12 v = [ 3 −1 ] −11 −12 3
−1 = 17.
h i
Hence kvk 1
v = √117 −13 is a unit vector in this space.
√
4. b. d(u, v) = ku − vk = k(1, 2, −1, 2) − (2, 1, −1, 3)k = k(−1, 1, 0, −1)k = 3.

151
152 Inner Product Spaces

d. k f − gk2 = −ππ (1 − cos x)2 dx = −ππ 32 − 2 cos x + 21 cos(2x) dx because we have cos2 (x) =
R R  
1 2 3 1
π
2 [1 + cos(2x)]. Hence k f − gk = 2 x − 2 sin(x) + 4 sin(2x) −π
= 32 [π − (−π )] = 3π . Hence
√
d( f , g) = 3π .
8. The space Dn uses pointwise addition and scalar multiplication:
( f + g)(k) = f (k) + g(k) and (r f )(k) = r f (k)
for all k = 1, 2, . . . , n.
P1. h f , gi = f (1)g(1) + f (2)g(2) + · · · + f (n)g(n) is real.
P2. h f , gi = f (1)g(1) + f (2)g(2) + · · ·+ f (n)g(n)) = g(1) f (1) + g(2) f (2) + · · ·+ g(n) f (n)
= hg, f i
P3. h f + h, gi = ( f + h)(1)g(1) + ( f + h)(2)g(2) + · · · + ( f + h)(n)g(n)
= [ f (1) + h(1)]g(1) + [ f (2) + h(2)]g(2) + · · ·+ [ f (n) + h(n)]g(n)
= [ f (1)g(1) + f (2)g(2) + · · ·+ f (n)g(n)] + [h(1)g(1) + h(2)g(2) + · · ·+ h(n)g(n)]
= h f , gi + hh, gi
P4. hr f , gi = (r f )(1)g(1) + (r f )(2)g(2) + · · ·+ (r f )(n)g(n)
= [r f (1)]g(1) + [r f (2)]g(2) + · · ·+ [r f (n)]g(n)
= r[ f (1)g(1) + f (2)g(2) + · · ·+ f (n)g(n)] = rh f , gi
P5. h f , f i = f (1)2 + f (2)2 + · · · + f (n)2 ≥ 0 for all f . If h f , f i = 0 then
f (1) = f (2) = · · · = f (n) = 0 (as the f (k) are real numbers) so f = 0

12. b. We need only verify P5. [P1 −h P4 ihold for any symmetric matrix A by (the discussion preced-
ing) Theorem 10.1.2.] If v = vv12 :
 h 5 −3 i h v1 i
hv, vi = vT Av = v1 v2

−3 2 v2

= 5v21 − 6v1 v2 + 2v22

= 5 v21 − 56 v1 v2 + 25
9 2
v2 − 95 v22 + 2v22
 

2
= 5 v1 − 35 v2 + 15 v22
h i
1 2 2
= 5 (5v1 − 3v2 ) + v2

Thus, hv, vi ≥ 0 for all v; and hv, vi = 0 if and only if 5v1 − 3v2 = 0 = v2 ; that is if and only
if v1 = v2 = 0 (i.e. v = 0). So P5 holds.
h i
d. As in (b), consider v = vv12 .
  h 3 4 i h v1 i
hv, vi = v1 v2 4 6 v2

= 3v21 + 8v1 v2 + 6v22

= 3(v21 + 83 v1 v2 + 16 2 16 2 2
9 v2 ) − 3 v2 + 6v2

2
= 3 v1 + 43 v2 + 23 v22
h i
= 31 (3v1 + 4v2 )2 + 2v22

Thus, hv, vi ≥ 0 for all v; and (v, v) = 0 if and only if 3v1 + 4v2 = 0 = v2 ; that is if and only
if v = 0. Hence P5 holds. The other axioms hold because A is symmetric.
 10.1. Inner Products and Norms 153

h i
a11 a12
13. b. If A = a21 , then ai j is the coefficient of vi w j in hv, wi. Here a11 = 1, a12 = −1 = a21 ,
a22
h i
1 −1
and a22 = 2. Thus, A = −1 2 . Note that a12 = a21 , so A is symmetric.
 
1 0 −2
d. As in (b): A = 0 2 0 .
−2 0 5

14. As in the hint, write hx, yi = xT Ay. Since A is symmetric, this satisfies axioms P1, P2, P3 and P4
for an inner product on Rn —(and only P2 requires that A be symmetric). Then it follows that

0 = hx + y, x + yi = hx, xi + hx, yi + hy, xi + hy, yi = 2hx, yi for all x, y in Rn .

Hence hx, yi = 0 for all x and y in Rn . But if e j denotes column j of In , then hei , e j i = eTi Ae j is the
(i, j)-entry of A. It follows that A = 0.
16. b. hu − 2v − w, 3w − vi = 3hu, wi − 6hv, wi − 3hw, wi − hu, vi + 2hv, vi + hw, vi
= 3hu, wi − 5hv, wi − 3 kwk2 − hu, vi + 2 kvk2
= 3 · 0 − 5 · 3 − 3 · 3 − (−1) + 2 · 4
= −15
P2 P3 P2
20. (1) hu, v + wi = hv + w, ui = hv, ui + hw, ui = hu, vi + hu, wi
P2 P4 P2
(2) hv, rwi = hrw, vi = rhw, vi = rhv, wi
(1)
(3) By (1): hv, 0i = hv, 0+0i = hv, 0i + hv, 0i. Hence hv, 0i = 0. Now h0, vi = 0 by P2.
(4) If v = 0 then hv, vi = h0, 0i = 0 by (3). If hv, vi = 0 then it is impossible that v 6= 0 by P5, so
v = 0.
22. b. h3u − 4v, 5u + vi = 15hu, ui + 3hu, vi − 20hv, ui − 4hv, vi
= 15 kuk2 − 17hu, vi − 4 kvk2
d. ku + vk2 = hu + v, u + vi = hu, ui + u, vi + hv, ui + 4hv, vi
= kuk2 + 2hu, vi + kvk2
26. b. Here

W = w | w in R3 and v · w = 0


= {(x, y, z) | x − y + 2z = 0}
= {(s, s + 2t, t) | s, t in R}
= span B

where B = {(1, 1, 0), (0, 2, 1)}. Then B is the desired basis because B is independent
[In fact, if s(1, 1, 0) + t(0, 2, 1) = (s, s + 2t, t) = (0, 0, 0) then s = t = 0].
28. Write u = v − w; we show that u = 0. We are given that

hu, vi i = hv − w, vi i = hv, vi i − hw, vi i = 0

for each i. As V = span {v1 , . . . , vn }, write u = r1 v1 + · · · + rn vn , ri in R. Then

kuk2 = hu, ui = hu, r1 v1 + · · · + rn vn i

154 Inner Product Spaces

= r1 hu, v1 ) + · · · + rn hu, vn i
= r1 · 0 + · · · + rn · 0
=0
Thus, kuk = 0, so u = 0.
29. b. If u = (cos θ , sin θ ) in R2 (with the dot product), then kuk = 1. If v = (x, y) the Schwarz
inequality (Theorem 10.1.4) gives
hu, vi2 ≤ kuk2 kvk2 ≤ 1 · kvk2 = kvk2
This is what we wanted.

10.2 Orthogonal Sets of Vectors

     
1 −1 1
1. b. B is an orthogonal set because (writing f1 = 1 , f2 = 0 and f3 = −6 )
1 1 1
    
  2 0 1 −1   −1
he1 , e2 i = 1 1 1 0 1 0 0 = 3 1 3 0 =0
1 0 2 1 1
    
  2 0 1 1   1
hf1 , f3 i = 1 1 1 0 1 0 −6 = 3 1 3 −6 =0
1 0 2 1 1
    
  2 0 1 1   1
hf2 , f3 i = −1 0 1 0 1 0 −6 = −1 0 1 −6 =0
1 0 2 1 1

Thus, B is an orthogonal basis of V and the expansion theorem gives

hv, f1 i hv, f2 i hv, f3 i
v= 2 f1 + 2 f2 + f3
kf1 k kf2 k kf3 k2
3a+b+3c 3a−6b+3c
= 7 e1 + c−a 2 e2 + 42 e3
1
= 14 [(6a + 2b + 6c)e1 + (7c − 7a)e2 + (a − 2b + c)e3 ]
Dh i h iE
a b a′ b′
d. Observe first that , c′ d′ = aa′ +bb′ +cc′ +dd ′ . Now write B = {f1 , f2 , f3 , f4 }
h i c d h i h i h i
where f1 = 10 01 , f2 = 10 0
−1
0 1 0 1
, f3 = 1 0 , f4 = −1 0 . Then B is orthogonal be-
cause
hf1 , f2 i = 1 + 0 + 0 − 1 = 0 hf2 , f3 i = 0 + 0 + 0 + 0 = 0
hf1 , f3 i = 0 + 0 + 0 + 0 = 0 hf2 , f4 i = 0 + 0 + 0 + 0 = 0
hf1 , f4 i = 0 + 0 + 0 + 0 = 0 hf3 , f4 i = 0 + 1 + 0 + −1 = 0
The expansion theorem gives
hv, f1 i
v= f1 + hv, f22i f2 + hv, f32i f3 + hv, f42i f4
kf1 k2 kf2 k kf3 k kf4 k
a+d
f1 + 2 f2 + 2 f3 + b−c
a−d b+c





= 2 2 f4
 10.2. Orthogonal Sets of Vectors 155

2. b. Write b1 = (1, 1, 1), b2 = (1, −1, 1), b3 = (1, 1, 0). Note that in the Gram-Schmidt algorithm
we may multiply each ei by a nonzero constant and not change the subsequent ei . This avoids
fractions.

f1 = b1 = (1, 1, 1)
hb2 , e1 i
f2 = b2 − f1
ke1 k2
= (1, −1, 1) − 46 (1, 1, 1)
= 13 (1, −5, 1); use e3 = (1, −5, 1) with no loss of generality
hb3 , f1 i hb3 , f2 i
f3 = b3 − 2 f1 − f2
ke1 k ke2 k2

= (1, 1, 0) − 36 (1, 1, 1) − (−3)

(30) · (1, −5, 1)
1
= 10 [(10, 10, 9) − (5, 5, 5) + (1, −5, 1)]
1
= 5 (3, 0, −2); use f3 = (3, 0, −2) with no loss of generality

So the orthogonal basis is {(1, 1, 1), (1, −5, 1), (3, 0, −2)}.
Dh i h ′ ′ iE
a b
3. b. Note that c d , ac′ db′ = aa′ + bb′ + cc′ + dd ′ . For convenience write
h i h i h i h i
1 1 1 0 1 0 1 0
b1 = 0 1 , b2 = 1 1 , b3 = 0 1 , b4 = 0 0 . Then:
h i
1 1
f1 = b1 = 0 1

f2 = b2 − hb2 , f21 i f1
kf1 k
h i h i h i
−2
= 11 01 − 23 1
0
1
1 = 1
3
1
3 1

h i
1 −2
For the rest of the algorithm, use f2 = 3 1 , the result is the same.

f3 = b3 − hb3 , f21 i f1 − hb3 , f22 i f2

kf1 k kf2 k
h i h i h i
−2
= 10 01 − 23 10 11 − 15 2 1
3 1
h i
1 1 −2
= 5 −2 1
h i
1 −2
Now use f4 = −2 1 , the results are unchanged.

hb , f i hb4 , f2 i hb , f i
f4 = b4 − 4 21 f1 − f2 − 4 23 f3
kf1 k kf2 k2 kf3 k
h i h i h i h i
−2 −2
= 10 00 − 31 10 1
1
1
− 15 1
3 1
1
− 10 1
−2 1
h i
1 1 0
= 2 0 −1
h i
1 0
Use f4 = 0 −1 for convenience. Hence, finally, the Gram-Schmidt algorithm gives the
nh i h i h i h io
1 1 1 −2 1 −2 1 0
orthogonal basis 0 1 , 3 1 , −2 1 , 0 −1 .
156 Inner Product Spaces

4. b. f1 = 1
hx, f1 i 2
f2 = x − 2 f1 = x − 2 · 1 = x − 1
kf1 k
2 2
f3 = x − hx , f21 i f1 − hx , f22 i f2
2 = x2 − 8/3 4/3 2 2
2 · 1 − 2/3 · (x − 1) = x − 2x + 3 .
kf1 k kf2 k

is in U ⊥ if and only if
 
6. b. x y z w
   
x+y = x y z w · 1 1 0 0 = 0

Thus y = −x and

U⊥ =
 
x −x z w | x, z, w in R
     
= span 1 −1 0 0 , 0 0 1 0 , 0 0 0 1

Hence dim U ⊥ = 3 and U = 1.

d. If p(x) = a + bx + cx2 , p is in U ⊥ if and only if
Z 1
0 = hp, xi = (a + bx + cx2 )xdx = a2 + b3 + 4c
0

Thus a = 2s + t, b = −3s, c = −2t where s and t are in R, so p(x) = (2s + t) − 3sx − 2tx2 .
Hence, U ⊥ = span 2 − 3x, 1 − 2x2 and dim U ⊥ = 2, dim U = 1.

 
a b
f. is in U if and only if
c d
Dh i h iE
a b 1 1
0= c d , 0 0 = a+b
Dh i h iE
0= a b
c d , 11 00 = a+c
Dh i h iE
0= a b
c d , 11 01 = a+c+d
nh i o nh io
a −a 1 −1
The solution d = 0, b = c = −a, so U ⊥ = −a 0 a in R = span −1 0 . Thus
dim U ⊥ = 1 and dim U = 3.
h i h i h i
7. b. Write b1 = 10 01 , b2 = 11 −11 , and b3 = 10 10 . Then {b1 , b2 , b3 } is independent but
not orthogonal. The Gram-Schmidt algorithm gives
h i
f1 = b1 = 10 01
h i h i h i
f2 = b2 − hb2 , f21 i f1 = 11 −11 − 02 10 01 = 11 −11
kf1 k
hb3 , f1 i hb , f i
f3 = b3 − f − 3 22 f2
kf1 k2 1 kf2 k
h i h i h i
1 1 1 1 0 2 1 1
= 0 0 −2 0 1 − 4 1 −1
h i
= 12 −10 10
 10.2. Orthogonal Sets of Vectors 157

h i h i
0 1 2 1
If E3′ = then E1 , E2 , E3′ is an orthogonal basis of U . If A =

−1 0 3 2 then
hA, E1 i hA, E2 i hA, E3′ i ′
projU A = 2 E1 + 2 E2 + 2E
kE1 k kE2 k kE3′ k 3
h i h i h i
4 1 0 4 1 1 −2 0 1
= 2 0 1 + 4 1 −1 + 2 −1 0
h i
3 0
= 2 1

is the vector in U closest to A.

8. b. We are given U = span {1, 1 + x2 }, and applying the Gram-Schmidt algorithm gives an or-
thogonal basis consisting of 1 and
2, 2 )1+(1+12 )1+(1+22 )1
(1 + x2 ) − h1+x 2
1i
1 = (1 + x2 ) − (1+0 1+1+1 = − 53 + x2
k1k

We use U = span {1, 5 − 3x2 }. Then Theorem 10.2.8 asserts that the closest vector in U to x
is
hx, 1i hx, 5−3x2 i 2 3 −12 2 3 2
projU x = 2 1+ 2 (5 − 3x ) = 3 + 78 (5 − 3x ) = 13 (1 + 2x )
k1k k5−3x2 k
Here, for example hx, 5 − 3x2 i = 0(5) + 1(2) + 2(−7) = −12, and the other calculations are
similar.
R1
9. b. {1, 2x − 1} is an orthogonal basis of U because h1, 2x − 1i = 0 (2x − 1)dx = 0. Thus
hx2 +1, 1i hx2 +1, 2x−1i
projU (x2 + 1) = 2 1+ (2x − 1)
k1k k2x−1k2
3/4 1/6
= 1 1 + 1/3 (2x − 1)
= x + 56
Hence, x2 + 1 = (x + 56 ) + (x2 − x + 16 ) is the required decomposition. Check: x2 − x + 16 is in
U ⊥ because
Z 1
2 1
x2 − x + 16 dx = 0


hx − x + 6 , 1i =
0
Z 1
hx2 − x + 16 , 2x − 1i = x2 − x + 16 (2x − 1)dx = 0


0

11. b. We have hv + w, v − wi = hv, vi − hv, wi + hw, vi − hu, ui = kvk2 − kwk2 . But this means
that hv + w, v − ui = 0 if and only if kvk = kwk. This is what we wanted.
14. b. If v is in U ⊥ then hv, ui = 0 for all u in U . In particular, hv, ui i = 0 for 1 ≤ i ≤ n, so v is in
{u1 , . . . , um }⊥ . This shows that U ⊥ ⊆ {u1 , . . . , um }⊥ . Conversely, if v is in {u1 , . . . , um }⊥
then hv, ui i = 0 for each i. If u is in U , write u = r1 u1 + · · · + rm um , ri in R. Then
hv, ui = hv, r1 u1 + · · · + rm um i
= r1 hv, u1 i + · · · + rm hv, um i
= r1 · 0 + · · · + rm · 0
=0
As u was arbitrary in U , this shows that v is in U ⊥; that is {u1 , . . . , um }⊥ ⊆ U ⊥ .
158 Inner Product Spaces

18. b. Write e1 = (3, −2, 5) and e2 = (−1, 1, 1), write B = {e1 , e2 }, and write U = span B. Then
B is orthogonal and so is an orthogonal basis of U . Thus if v = (−5, 4, −3) then
v·e1
projU v = e1 + v·e22 e2
ke1 k2 ke2 k
−38 6
= 38 (3, −2, 5) + 3 (−1, 1, 1)
= (−5, 4, −3)
=v

Thus, v is in U . However, if v1 = (−1, 0, 2) then

v1 ·e1
projU v1 = e1 + v·e22 e2
ke2 k2 ke2 k
7 3
= 38 (3, −2, 5) + 3 (−1, 1, 1)
1
= 38 (−17, 24, 73)

As v1 6= projU v1 , v1 is not in U by (a).

n   o
n·w
19. b. The plane is U = {x | x · n = 0}, so span n × w, w − knk 2 n ⊆ U . Since dim U = 2,
n   o
n·w
it suffices to show that B = n × w, w − knk 2 n is independent. These two vectors are
orthogonal (because (n×w) · n = 0 = (n×w) · w). Hence B is orthogonal (and so independent)
provided each of the vectors is nonzero. But: n × w 6= 0 because n and w are not parallel, and
n·w n·w
w − knk 2 n is nonzero because w and n are not parallel, and n · (w − knk2 n) = 0.

20. b. CE (bi ) is column i of P. Since CE (bi ) ·CE (b j ) = hbi , b j i by (a), the result follows.
23. b. Let V be an inner product space, and let U be a subspace of V . If U = span {f1 , . . . , fm }, then
m m 2
projU v = ∑ hv,kfi kfi2i fi by Theorem 10.2.7 so k projU vk2 = ∑ hv,kfifki2i by Pythagoras’ theorem.
i=1 i=1
So it suffices to show that k projU vk2 ≤ kvk2 .
Given v in V , write v = u + w where u = projU v is in U and w is in U ⊥. Since u and w are
orthogonal, Pythagoras’ theorem (again) gives

kvk2 = kuk2 + kwk2 ≥ kuk2 = k projU vk2

This is what we wanted.

10.3 Orthogonal Diagonalization

h i h i h i
1. b. If B = {E1 , E2 , E3 , E4 } where E1 = 10 00 , E2 = 00 10 , E3 = 0
1
0
0
h i
and E4 = 00 01 , then B is an orthonormal basis for M22 and
h i
−1 0
T (E1 ) = 1 0 = −E1 + E3
 10.3. Orthogonal Diagonalization 159

h i
0 −1
T (E2 ) = 0 1 = −E2 + E4
h i
1 0
T (E3 ) = 2 0 = E1 + 2E3
h i
0 1
T (E4 ) = 0 2 = E2 + 2E4

Hence,
 
MB (T ) = CB [T (E1 )] CB [T (E2 )] CB [T (E3 )] CB [T (E4 )]
" −1 0 1 0 #
0 −1 0 1
= 1 0 2 0
0 1 0 2

As MB (T ) is symmetric, T is a symmetric operator.

4. b. If T is symmetric then hv, T (w)i = hT (v), wi holds for all v and w in V . Given r in R:

hv, (rT )(w)i = hv, rT (w)i = rhv, T (w)i = rhT (v), wi = hrT (v), wi = h(rT )(v), wi

for all v and w in V . This shows that rT is symmetric.

d. Given v and w, write T −1 (v) = v1 and T −1 (w) = w1 . Then

hT −1 (v), wi = hv1 , T (w1 )i = hT (v1 ), w1 i = hv, T −1 (w)i

This shows that T −1 is a symmetric operator.

5. b. If E = {e1 = (1, 0, 0), e2 = (0, 1, 0), e3 = (0, 0, 1)} is the standard basis of R3 :
 
ME (T ) = CE [T (e1 )] CE [T (e2 )] CE [T (e3 )]
 
= CE (7, −1, 0) CE (−1, 7, 0) CE (0, 0, 2)
 
7 −1 0
= −1 7 0
0 0 2

x−7 1 0
Thus, cT (x) = 1 x−7 0 = (x − 6)(x − 8)(x − 2) so the eigenvalues are λ1 = 6, λ2 = 8,
0 0 x−2
and
  λ3 = 2, (real as
 MB0 (T ) issymmetric).
 Corresponding (orthogonal) eigenvectors are x1 =
1 1 0
1 , x2 = −1 , and x3 = 0 , so
0 0 1
      
1 1 0
√1 1 , √1 −1 , 0
2 0 2 0 1
h i
is an orthonormal basis of eigenvectors of ME (T ). These vectors are equal to CE √1 (1, 1, 0) ,
h i 2
1
CE 2 (1, −1, 0) , and CE [(0, 0, 1)] respectively, so
√

n o
√1 (1, 1, 0), √1 (1, −1, 0), (0, 0, 1)
2 2

is an orthonormal basis of eigenvectors of T .

160 Inner Product Spaces

d. If B0 = 1, x, x2 then


CB0 [T (1)] CB0 [T (x)] CB0 [T (x2 )]

 
MB0 (T ) =
CB0 (−1 + x2 ) CB0 (3x) CB0 (1 − x2 )
 
=
 
−1 0 1
= 0 3 0
1 0 −1

x+1 0 −1
Hence, cT (x) = 0 = x(x−3)(x+2) so the (real) eigenvalues are λ1 = 3, λ2 = 0,
x−3 0
−1 0 x+1
     
0 1 1
λ3 = −2. Corresponding (orthogonal) eigenvectors are x1 = 1 , x2 = 0 , x3 = 0 ,
0 1 −1
     
0 1 1
so 1 , √1 0 , √1 0 is an orthonormal basis of eigenvectors of MB0 (T ). These
0 2 1 2 −1
h i h i
have the form CB0 (x), CB0 √1 (1 + x2 ) , and CB0 √1 (1 − x2 ) , respectively, so
2 2
n o
x, √1 (1 + x2 ), √1 (1 − x2 )
2 2
is an orthonormal basis of eigenvectors of T .
h i
7. b. Write A = ac db and compute:
h  h i  h i  h i  h i i
MB (T ) = CB T 10 00 CB T 01 00 CB T 0
0
1
0 CB T 0
0
0
1
h h i h i h i h i i
a 0 b 0 0 a 0 b
= CB c 0 CB d 0 CB 0 c CB 0 d
" a b 0 0 #
c d 0 0
= 0 0 a b
0 0 c d
h i
A 0
= 0 A

Hence,
nh i h io
xI 0 A 0
cT (x) = det [xI − MB (T )] = det − 0 xI 0 A
h i
= det xI −0 A xI −0 A = det (xI − A) · det (xI − A) = [cA (x)]2

12. (2) We prove that (1) ⇒ (2). If B = {f1 , . . . , fn } is an orthonormal basis of V , then MB (T ) = [ai j ]
where ai j = hfi , T (f j )i by Theorem 10.3.2. If (1) holds then a ji = hf j , T (fi )i = −hT (f j ), fi i =
−hfi , T (f j )i = −ai j . Hence [MV (T )]T = −MV (T ), proving (2).
14. c. We have
MB (T ′ ) = CB [T ′ (f1 )] CB [T ′ (f2 )] · · · CB [T ′ (fn )]
 

Hence, column j of MB (T ′ ) is
hf j , T (f1 )i
 
hf j , T (f2 )i
CB (T ′ (f j )) =  .. 
.
hf j , T (fn )i

by the definition of T ′ . Hence the (i, j)-entry of MB (T ′ ) is hf j , T (fi )i. But this is the ( j, i)-
entry of MB (T ) by Theorem 10.3.2. Thus, MB (T ′ ) is the transpose of MB (T ).
 10.4. Isometries 161

10.4 Isometries
h i h i h ih i h i
2. b. We have T ab = −a −b = −1 0
0 −1
a
b so T has matrix −1 0
0 −1 , which is orthogonal.
Hence T is an isometry, and det T = 1 so T is a rotation by Theorem 10.4.4.
h In fact,i T
cos θ − sin θ
is counterclockwise rotation through π . (Rotation through θ has matrix sin θ cos θ by
Theorem 2.6.4; see also the discussion following Theorem 10.4.3). This can also be seen
directly from the diagram.

y
 
a
b

   
a −a
T =
b −b

h i h i h ih ii h
a −b 0 −1 a 0 −1
d. We have T b = = −a so T has matrix
−1 0 b . This is orthogonal, −1 0
so T is an isometry. Moreover, det T = −1 so T is a reflection by Theorem 10.4.4. In fact,
T is reflection in the line y = −x by Theorem 2.6.5. This can also be seen directly from the
diagram.

y
 
a
b

   
a −b
T = y = −x
b −a

f. If B0 is the standard basis of R2 , then

h  h i  h i i
MB0 (T ) = CB0 T 10 CB0 T 01
h  h i  h i i
1 1 1 −1
= CB0 √
2 1
CB0 √
2 1
h i
= √1 11 −11
2
162 Inner Product Spaces

Hence, det T = 1 so T is a rotation. Indeed, (the discussion following) Theorem 10.4.3 shows
that T is a rotation through an angle θ where cos θ = √1 , sin θ = √1 ; that is θ = π4 .
2 2
   √   √    √ 
a
1 √3c − a 1
−1
√ 0 3 a
1
−1
√ 0 3
3. b. T b = 2
3a + c = 2
3 0 1 b , so T has matrix 2
3 0 1 . Thus,
c 2b 0 2 0 c 0 2 0

√ √
3 3
x + 12 0 − 2 x+ 1
0 − √
√ √2 2
x+ 1
− 3
= (x − 1) x2 + 32 x + 1


cT (x) = − 23 x − 21 = − 23 0 x2 − 12 = √2
− 23
2
x − 12
2
0 −1 x 0 −1 x

Hence,
 1we are in (1) of Table 10.1 so T is a rotation about the line Re with direction vector
√
e = √3 , where e is an eigenvector corresponding to the eigenvalue 1.
3
        
a a 1 0 0 a 1 0 0
d. T b = −b = 0 −1 0 b , so T has matrix 0 −1 0 . This is orthogonal,
c −c 0 0 −1 c 0 0 −1
2
 isometry. Since cT (x) = (x − 1)(x + 1) , we are in case (4) of Table 10.1. Then
so Tis an
1
e = 0 is an eigenvector corresponding to 1, so T is a rotation of π about the line Re with
0
direction vector e, that is the x-axis.
        
a +c
a√ 1 0
√ 1 a 1 0
√ 1
1 1 √1
f. T b = 2 − 2b = 2
√ √ 0 − 2 0 b , so T has matrix 2
0 − 2 0 . Hence,
c c−a −1 0 1 c −1 0 1

x − √12 0 − √12 √
x − √1 − √1
cT (x) = 0 x+1 0 = (x + 1) √1
2 2
x − √12
= (x + 1)(x2 − 2x + 1)
√1 0 x − √1 2
2 2
 
0
Thus we are in case (2) of Table 10.1. Now e = 1 is an eigenvector corresponding to the
0
3π
eigenvalue −1, so T is rotation (of 4 ) about the line Re (the y-axis) followed by a reflection
in the plane (Re)⊥ — the xz-plane.

6. Let T be an arbitrary isometry, and let a be a real number. If aT is an isometry then Theorem 10.4.2
gives
kvk = k(aT )(v)k = ka(T (v))k = |a| kT (v)k = |a|kvk holds for all v.
Thus |a| = 1 so, since a is real, a = ±1. Conversely, if a = ±1 then |a| = 1 so we have k(aT )(v)k =
|a| kT (v)k = 1 kT (v)k = kvk for all v. Hence aT is an isometry by Theorem 10.4.2.

12. b. Assume that S = Su ◦ T where u is in V and T is an isometry of V . Since T is onto (by Theorem
10.4.2), let u = T (w) where w ∈ V . Then for any v ∈ V , we have

(T ◦ Sw )(v) = T (w + v) = T (w) + T (v) = ST (w) (T (v)) = (ST (w) ◦ T )(v)

Since this holds for all v ∈ V , it follows that T ◦ Sw = ST (w) ◦ T .

 10.5. An Application to Fourier Approximation 163

10.5 An Application to Fourier Approximation

The integrations involved in the computation of the Fourier coefficients are omitted in 1(b), 1(d), and 2(b).
 
π
1. b. f5 = 2 − π4 cos x + cos323x + cos525x
 
π
+ sin x − sin22x + sin33x − sin44x + sin55x − π2 cos x + cos323x + cos525x


d. f5 = 4
 
2
2. b. π − π8 cos 2x
22 −1
+ cos 4x
42 −1
+ cos 6x
62 −1

4. We use the formula that cos(θ ± φ ) = cos θ cos φ ∓ sin θ sin φ , so that 2 cos θ cos φ = cos(θ −
φ ) cos(θ + φ ). Hence:
Z π Z π
1
cos(kx) cos(ℓx)dx = 2 {cos[(k − ℓ)x] + cos[(k + ℓ)x]} dx
0 h0
sin[(k−ℓ)x] π
i
1 sin[(k+ℓ)x]
= 2 k+ℓ + (k−ℓ)
0
= 0 if k 6= ℓ.
11. Canonical Forms

11.1 Block Triangular Form

x+5 −3 −1 x+1 −x − 1 0 x+1 0 0

1. b. cA (x) = 4 x−2 −1 = 4 x−2 −1 = 4 x+2 −1 = (x + 1)3 .
4 −3 x 4 −3 x 4 1 x
Hence, λ1 = −1 and we are in case k = 1 of the triangulation algorithm.
       
4 −3 −1 4 −3 −1 1 0
−I − A = 4 −3 −1 → 0 0 0 ; p11 = 1 , p12 = 1
4 −3 −1 0 0 0 1 −3

2 .
 
Hence, {p11 , p12 } is a basis of null (−I −A). We now expand this to a basis of null (−I − A)
However, (−I − A)2 = 0 so null (−I − A)2 = R3 . Hence,
 
 this case, we expand {p11 , p12 }
 in
0
to any basis {p11 , p12 , p13 } of R3 , say by taking p13 = 0 . Hence
1
   
1 0 0 −1 0 1
−1
 
P= p11 p12 p13 = 1 1 0 satisfies P AP = 0 −1 0
1 −3 1 0 0 −1

as may be verified.
x+3 1 0 x+3 1 0 x+3 1 0
d. cA (x) = −4 x+1 −3 = −4 x+1 −3 = −4 x−2 −3 = (x − 1)2 (x + 2).
−4 2 x−4 0 −x + 1 x−1 0 0 x−1
Hence λ1 = 1, λ3 = −2, and we are in case k = 2 of the triangulation algorithm.
     
4 1 0 4 1 0 −1
I − A = −4 2 −3 → 0 3 −3 ; p11 = 4
−4 2 −3 0 0 0 4

Thus, null (I − A) = span {p11 }. We enlarge {p11 } to a basis of null (I − A)2

 

       
12 6 −3 4 2 −1 −1 0
12
(I − A) = −12 −6 3 → 0 0 0 ; p11 = 4 , p12 = 1
−12 −6 3 0 0 0 4 2

Thus, null (I − A)2 = span {p11 , p12 }. As dim Gλ1 (A) = 2 in this case (by Lemma 11.1.1),
   

we have Gλ1 (A) = span {p11 , p12 }. However, it is instructive to continue the process:
 
4 2 −1
2
(I − A) = 3 −4 −2 1
−4 −2 1

whence  
4 2 −1
3
(I − A) = 9 −4 −2 1 = 3(I − A)2
−4 −2 1

165
166 Canonical Forms

This continues tohgive (I −i A)4 = 3h2 (I − A)2i, . . . , and in general (I − A)k = 3k−2 (I − A)2 for
k ≥ 2. Thus null (I − A)k = null (I − A)2 for all k ≥ 2, so
h i
Gλ1 (A) = null (I − A)2 = span {p11 , p12 }

as we expected. Turning to λ2 = −2:

       
1 1 0 1 1 0 1 1 0 −1
−2I − A = −4 −1 −3 → 0 3 −3 → 0 1 −1 ; p21 = 1
−4 2 −6 0 6 −6 0 0 0 1

Hence, null [−2I − A] = span {p21 }. We need go no further with this as {p11 , p12 , p21 } is a
basis of R3 . Hence
   
−1 0 −1 1 1 0
−1
 
P = p11 p12 p21 = 4 1 1 satisfies P AP = 0 1 0
4 2 1 0 0 −2

as may be verified.
f. To evaluate cA (x), we begin by adding column 4 to column 1:
x+3 −6 −3 −2 x+1 −6 −3 −2 x+1 −6 −3 −2
2 x−3 −2 −2 0 x−3 −2 −2 0 x−3 −2 −2
cA (x) = 1 −3 x −1 = 0 −3 x −1 = 0 −3 x −1
1 −1 −2 x x+1 −1 −2 x 0 5 1 x+2
x−3 −2 −2 x−3 −2 0 x−3 −2 0
= (x + 1) −3 x −1 = (x + 1) −3 x −x − 1 = (x + 1) 2 x+1 0
5 1 x+2 5 1 x+1 5 1 x+1
x−3 −2
= (x + 1)1 2 x+1 = (x + 1)2 (x − 1)2

Hence, λ1 = −1, λ2 = 1 and we are in case k = 2 of the triangulation algorithm. We omit the
details of the row reductions:
" 2 −6 −3 −2 # " 1 0 0 −1 # " 1 #
2 −4 −2 −2 0 1 0 0 0
−I − A = 1 −3 −1 −1 → 0 0 1 0 ; p11 = 0
1 −1 −2 −1 0 0 0 0 1
−13 23 13 13 1 0 −1 −1 1 1
" # " # " # " #
−8 12 8 8 0 1 0 0 0 0
(−I − A)2 = −6 10 6 6 → 0 0 0 0 ; p11 0 , p12 = 1
−3 5 3 3 0 0 0 0 1 0
 
We have dim Gλ1 (A) = 2 as λ1 = −1 has multiplicity 2 in cA (x), so Gλ1 (A) = span {p11 , p12 }.
Turning to λ2 = 1:
" 4 −6 −3 −2 # " 1 0 0 −5 # " 5 #
2 −2 −2 −2 0 1 0 −2 2
I −A = 1 −3 1 −1 → 0 0 1 −2 ; p21 = 2
1 −1 −2 1 0 0 0 0 1
−1 −1 1 5 1 1 0 0 5 1
" # " # " # " #
0 0 0 0 0 0 1 0 2 −1
(I − A)2 = −2 −2 2 −6 → 0 0 0 1 ; p21 = 2 , p22 = 0
1 1 −5 3 0 0 0 0 1 0

Hence, Gλ2 (A) = span {p21 , p22 } using Lemma 11.1.1. Finally, then
−1
" 1 1 5 1 # " 1 0 0
#
0 0 2 −1 0 −1 0 0
P = p11 p12 p21 p22 = 0 1 2 0 gives P−1 AP =
 
0 0 1 −2
1 0 1 0 0 0 0 1

as may be verified.
 11.2. Jordan Canonical Form 167

4. Let B be any basis of V and write A = MB (T ). Then cT (x) = cA (x) and this is a polynomial: cT (x) =
a0 + a1 x + · · · + an xn for some ai in R. Now recall that MB : L(V , V ) → Mnn is an isomorphism
of vector spaces (Exercise 9.1.26) with the additional property that MB (T k ) = MB (T )k for k ≥ 1
(Theorem 9.2.1). With this we get

MB [cT (T )] = MB [a0 1V + a1 T + · · · + an T n ]
= a0 MB (1V ) + a1 MB (T ) + · · · + an MB (T )n
= a 0 I + a 1 A + · · · + a n An
= cA (A)
=0

by the Cayley-Hamilton theorem. Hence cT (T ) = 0 because MB is one-to-one.

11.2 Jordan Canonical Form

       
a 1 0 0 1 0 0 1 0 b 0 0 0 1 0
2. 0 a 0 0 0 1 = 0 0 1 0 a 1 , and 0 0 1 is invertible.
0 0 b 1 0 0 1 0 0 0 0 a 1 0 0
A. Complex Numbers
1. b. 12 + 5i = (2 + xi)(3 − 2i) = (6 + 2x) + (−4 + 3x)i. Equating real and imaginary parts gives
6 + 2x = 12, −4 + 3x = 5, so x = 3.
d. 5 = (2 + xi)(2 − xi) = (4 + x2 ) + 0i. Hence 4 + x2 = 5, so x = ±1.
√
2. b. (3 − 2i)(1 + i) + |3 + 4i| = (5 + i) + 9 + 16 = 10 + i
3−2i (3−2i)(1+i) (3−7i)(2+3i)
d. 1−i − 3−7i
2−3i = (1−i)(1+i) − (2−3i)(2+3i)
5+i 27−5i
= 1+1 − 4+9
11 23
= 26 + 26 i
f. (2 − i)3 = (2 − i)2(2 − i) = (3 − 4i)(2 − i) = 2 − 11i
h. (1 − i)2 (2 + i)2 = (−2i)(3 + 4i) = 8 − 6i

3. b. iz + 1 = i + z − 6i + 3iz = −5i + (1 + 3i)z. Hence 1 + 5i = (1 + 2i)z, so

1+5i (1+5i)(1−2i) 11+3i 11
z= 1+2i = (1+2i)(1−2i) = 1+4 = 5 + 35 i

d. z2 = 3 − 4i. If z = a + bi the condition is (a2 − b2 ) + (2ab)i = 3 − 4i, whence a2 − b2 =

3 and ab = −2. Thus b = −2 2 4 4 2
a , so a − a2 = 3. Hence a − 3a − 4 = 0. This factors as
(a2 − 4)(a2 + 1) = 0, so a = ±2, whence b = ∓1. Finally, z = a + bi = ±(2 − i).
f. Write z = a + bi. Then the condition reads

(a + bi)(2 − i) = (a − bi + 1)(1 + i)
(2a + b) + (2b − a)i = (a + 1 + b) + (a + 1 − b)i

Thus 2a + b = a + 1 + b and 2b − a = a + 1 − b; whence a = 1, b = 1, so z = 1 + i.

h i √ 
b. x = 21 −(−1) ± (−1)2 − 4 = 12 1 ± i 3
p 
4.
h i √ 
d. x = 41 −(−5) ± (−5)2 − 4 · 2 · 2 = 14 5 ± 9 = 2, 12
p 

5. b. If x = reiθ then x3 = −8 becomes r3 e3iθ = 8eπ i . Thus r3 = 8 (whence r = 2) and 3θ = π +2kπ .

Hence θ = π3 + k · 23π , k = 0, 1, 2. The roots are
√
2eiπ /3 = 1 + 3i (k = 0)
2eπ i = −2 (k = 1)
√
2e5π i/3 = 1 − 3i (k = 2)

169
170 Complex Numbers

√
d. If x = re−iθ then x4 = 64 becomes r4 e4iθ = 64ei·0 . Hence r4 = 64 (whence r = 2 2) and
4θ = 0 + 2kπ ; θ = k π2 , k = 0, 1, 2, 3. The roots are
√ √
2 2e0i = 2 2 (k = 0)
√ √
2 2eπ i/2 = 2 2i (k = 1)
√ √
2 2eπ i = −2 2 (k = 2)
√ √
2 2e3π i/2 = −2 2i (k = 3)

6. b. The quadratic is (x−u)(x−u) = x2 −(u +u)x+u u = x2 −4x+13. The other root is u = 2 +3i.
d. The quadratic is (x−u)(x−u) = x2 −(u +u)x+u u = x2 −6x+25. The other root is u = 3 +4i.

8. If u = 2 − i, then u is a root of (x − u)(x − u) = x2 − (u + u)x + uu = x2 − 4x + 5.

If v = 3 − 2i, then v is a root of (x − v)(x − v) = x2 − (v + v)x + vv = x2 − 6x + 13.
Hence u and v are roots of

(x2 − 4x + 5)(x2 − 6x + 13) = x4 − 10x3 + 42x2 − 82x + 65

10. b. Taking x = u = −2: x2 +ix −(4 −2i) = 4 −2i −4 +2i = 0. If v is the other root then u +v = −i
(i is the coefficient of x) so v = −u − i = 2 − i.
d. Taking x = u = −2 + i: (−2 + i)2 = 3(1 − i)(−2 + i) − 5i
= (3 − ri) + 3(−1 + 3i) − 5i
= 0.
If v is the other root then u + v = −3(1 − i), so v = −3(1 − i) − u = −1 + 2i.
h i √ √
b. x2 − x + (1 − i) = 0 gives x = 21 1 ± 1 − 4(1 − i) = 21 1 ± −3 + 4i . Write w = −3 + 4i
p  
11.
so w2 = −3 + 4i. If w = a + bi then w2 = (a2 − b2 ) + (2ab)i, so a2 − b2 = −3, 2ab = 4. Thus
b = 2a , a2 − a42 = −3, a4 + 3a2 − 4 = 0, (a2 + 4)(a2 − 1) = 0, a = ±1, b = ±2, w = ±(1 + 2i).
Finally the roots are 12 [1 ± w] = 1 + i, −i.
h i √ 
d. x − 3(1 − i)x − 5i = 0 gives x = 2 3(1 − i) ± 9(1 − i) + 20i = 12 3(1 − i) ± 2i . If w =
1
p
2

2
√
2i then w2 = 2i. Write w = a + bi so (a2 − b2 ) + 2abi = 2i. Hence a2 = b2 and ab = 1; the
solution is a = b = ±1 so w = ±(1 + i). Thus the roots are x = 12 (3(1 − i) ± w) = 2 − i, 1 − 2i.

12. b. |z − 1| = 2 means that the distance from z to 1 is 2. Thus the graph is the circle, radius 2, center
at 1.
d. If z = x + yi, then z = −z becomes x + yi = −x + yi. This holds if and only if x = 0; that is if
and only if z = yi. Hence the graph is the imaginary axis.
f. If z = x + yi, then im z = m · re z becomes y = mx. This is the line through the origin with slope
m.

18. b. −4i = 4e3π i/2

√ √
d. −4 + 4 3i = 4 1 + 3 = 8 and cos ϕ = 4
= 12 . Thus ϕ = π3 , so θ = 2π
and we have −4 +
√ 8 3
4 3i = 8e2π i/3 .
 171

√ √
f. |−6 + 6i| = 6 1 + 1 = 6 2 and cos ϕ = 6√6 2 = √12 . Thus ϕ = π4 so θ = 34π ; whence −6 + 6i =
√
6 2e−3π i/4 .
√
19. b. e7π i/3 = e(π /3+2π )i = eπ i/3 = cos π3 + i sin π3 = 12 + 23 i
√ −π i/4 √ −π

−π


√  1 1

d. 2e = 2 cos 4 + i sin 4 = 2 √ − i = 1−i
√
2 2
√ −2π i/6 √ √  √  √
= 2 3 cos −3π + i sin −3π = 2 3 12 − 23 i = 3 − 3i




f. 2 3e
√
20. b. (1 + 3i)−4 = (2eπ i/3 )−4 = 2−4 e−4π i/3
1
= 16 [cos(−4
 π /3) + i sin(−4π /3)]
√ 
= 16 − 2 + 23 i
1 1
√
= − 321
+ 323 i
h√ i10 √ √
d. (1 − i)10 = 2e−π i/4 = ( 2)10 e−5π i/2 = ( 2)10 e(−π /2−2π )i
√
= ( 2)10 e−π i/2 = 25 cos −2π + i sin −2π




= 32(0 − i) = −32i
√
 h √ i5 
9 5
f. ( 3 − i) (2 − 2i) = 2e − π i/6 9
] 2 2e −π i/4
√
= 29 e−3π i/2 (2 2)5 e−5π i/4
√ √ 
= 29 (i)25 ( 2)4 2 − √12 + √12 i
= 216 i(−1 + i)
= −216 (1 + i)
√ √
23. b. Write z = reiθ . Then z4 = 2( 3i − 1) becomes r4 e4iθ = 4e2π i/3 . Hence r4 = 4, so r = 2, and
4θ = 23π + 2π k; that is
θ = π6 + π2 k k = 0, 1, 2, 3
The roots are
√ √  √3 1  √2 √ 
2eπ i/6 =
2 2 + 2i = 2 3+i
√ 4π i/6 √  √ 
3
√ 
2
√ 
1
2e = 2 − 2 + 2 i = 2 −1 + 3i
√ 7π i/6 √  √3 1  √ √ 
2e = 2 − 2 − 2 i = − 22 3+i
√ 10π i/6 √  1 √3  √  √ 
2e = 2 2 − 2 i = − 22 −1 + 3i
172 Complex Numbers

k=2

k=1

π /6

k=3
k=4

d. Write z = reiθ . Then z6 = −64 becomes r6 e6iθ = 64eπ i . Hence r6 = 64, so r = 2, and 6θ =
π + 2π k; that is θ = π6 + π3 k where k = 0, 1, 2, 3, 4, 5. The roots are thus z = 2eπ /6+π /3k for
these values of k. In cartesian form they are

k √0 1 √2 √3 4 √5
z 3 + i 2i − 3 + i − 3 − i −2i 3−i

26. b. Each point on the unit circle has polar form eiθ for some angle θ . As the n points are equally
spaced, the angle between consecutive points is 2nπ . Suppose the first point into the first quad-
rant is z0 = eα i . Write w = e2π i/n . If the points are labeled z1 , z2 , z3 , . . . , zn around the unit
circle, they have polar form

z1 = eα i
z2 = e(α +2π /n)i = eα i e2π i/n = z1 w
z3 = e[α +2(2π /n)]i = eα i e4π i/n = z1 w2
z4 = e[α +3(2π /n)]i = eα i e6π i/n = z1 w3
..
.
zn = e[α +(n−1)(2π /n)]i = eai e2(n−1)π i/n = z1 wn−1

z4 z3

z2
2π
n 2π
n
2π
n z1
α
 173

Hence the sum of the roots is

z1 + z2 + · · · + zn = z1 (1 + w + · · · + wn−1 ) (∗)
 n
Now wn = e2π i/n = e2π i = 1 so

0 = 1 − wn = (1 − w)(1 + w + w2 + · · · + wn−1 )

As w 6= 1, this gives 1 + w + · · · + wn−1 = 0. Hence (∗) gives

z1 + z2 + · · · + zn = z1 · 0 = 0
B. Proofs
1. b. (1). We are to prove that if the statement “m is even and n is odd ” is true then the statement
“m + n is odd ” is also true.
If m is even and n is odd, they have the form m = 2p and n = 2q + 1, where p and q are
integers. But then m + n = 2(p + q) + 1 is odd, as required.
(2). The converse is false. It states that if m + n is odd then m is even and n is odd; and a
counterexample is m = 1, n = 2.
d. (1). We are to prove that if the statement “x2 − 5x + 6 = 0” is true then the statement “x = 2 or
x = 3” is also true.
Observe first that x2 −5x+6 = (x−2)(x−3). So if x is a number satisfying x2 −5x+6 = 0
then (x − 2)(x − 3) − 0 so either x = 2 or x = 3. [Note that we are using an important fact
about real numbers: If the product of two real numbers is zero then one of them is zero.]
(2). The converse is true. It states that if x = 2 or x = 3 then x satisfies the equation x2 −5x+6 =
0. This is indeed the case as both x = 2 or x = 3 satisfy this equation.

2. b. The implication here is p ⇒ q where p is the statement “n is any odd integer ”, and q is the
statement “n2 = 8k + 1 for some integer k”. We are asked to either prove this implication or
give a counterexample.
This implication is true. If p is true then n is odd, say n = 2t + 1 for some integer t. Then
n2 = (2t)2 + 2(2t) + 1 = 4t(t + 1) + 1. But t(t + 1) is even (because t is either even or odd),
say t(t + 1) = 2k where k is an integer. Hence n2 = 4t(t + 1) + 1 = 4(2k) + 1, as required.

3. b. The implication here is p ⇒ q where p is the statement “n + m = 25, where n and m are
integers ”, and q is the statement “one of m and n is greater than 12” is also true. We are asked
to either prove this implication by the method of contradiction, or give a counterexample.
The implication is true. To prove it by contradiction, we assume that the conclusion q is false,
and look for a contradiction. In this case assuming that q is false means both n ≤ 12 and
m ≤ 12. But then n + m ≤ 24, contradicting the hypothesis that n + m = 25. So the statement
is true by the method of proof by contradiction.
The converse is false. It states that q ⇒ p, that is if one of m and n is greater than 12 then
n + m = 25. But n = 13 and m = 13 is a counterexample.
d. The implication here is p ⇒ q where p is the statement “mn is even, where n and m are
integers ”, and q is the statement “m is even or n is even ”. We are asked to either prove this
implication by the method of contradiction, or give a counterexample.
This implication is true. To prove it by contradiction, we assume that the conclusion q is false,
and look for a contradiction. In this case assuming that q is false means that m and n are both
odd. But then mn is odd (if either were even the product would be even). This contradicts the
hypothesis, so the statement is true by the method of proof by contradiction.
The converse is true. It states that if m or n is even then mn is even, and this is true (if m or n
is a multiple of 2, then mn is a multiple of 2).

175
176 Proofs

4. b. The implication here is: “x is irrational and y is rational ” ⇒ “x + y is irrational ”.

To argue by contradiction, assume that x + y is rational. Then x = (x + y) − y is the difference
of two rational numbers, and so is rational, contrary to the hypothesis that x is irrational.

5. b. At first glance the statement does not appear to be an implication. But another way to say it is
that if the statement “n ≥ 2” is true then the statement “n3 ≥ 2n ” is also true.
This is not true. In fact, n = 10 is a counterexample because 103 = 1000 while 210 = 1024. It
is worth noting that the statement n3 ≥ 2n does hold for 2 ≤ n < 9.
 C. Mathematical Induction

6. Write Sn for the statement

1 1 1 n
1·2 + 2·3 + · · · + n(n+1) = n+1 (Sn )
1 1
Then S1 is true: It reads 1·2 = 1+1 , which is true. Now assume Sn is true for some n ≥ 1. We must
use Sn to show that Sn+1 is also true. The statement Sn+1 reads as follows:
1 1 1 n+1
1·2 + 2·3 + · · · + (n+1)(n+2) = n+2

1
The second last term on the left side is n(n+1)
so we can use Sn :
h i
1 1 1 1 1 1 1
1·2 + 2·3 + · · · + (n+1)(n+2) = 1·2 + 2·3 + · · · + n(n+1) + (n+1)(n+2)
n 1
= n+1 + (n+1)(n+2)
n(n+2)+1
= (n+1)(n+2)
(n+1)2
= (n+1)(b+2)
n+1
= n+2

Thus Sn+1 is true and the induction is complete.

14. Write Sn for the statement √

√1 + √1 + · · · + √1n ≤ 2 n − 1 (Sn )
1 2
√
Then S1 is true as it asserts that √1 ≤ 2 1 − 1, which is true. Now assume that Sn is true for some
1
n ≥ 1. We must use Sn to show that Sn+1 is also true. The statement Sn+1 reads as follows:
h i
√1 + √1 + · · · + √ 1 = √1 + √1 + · · · + √1 + √ 1
1 2 n+1 1 2 n n+1
 √  1
≤ 2 n − 1 + √n+1
√
2 √n2 +n+1
= n+1
−1
2(n+1)
< √
n+1
−1
√
= 2 n+1−1
√
where, at the second last step, we used the fact that n2 + n < (n + 1)—this follows by showing that
n2 + n < (n + 1)2 , and taking positive square roots. Thus Sn+1 is true and the induction is complete.

18. Let Sn stand for the statement

n3 − n is a multiple of 3.

177
178 Mathematical Induction

Clearly S1 is true. If Sn is true, then n3 − n = 3k for some integer k. Compute:

(n + 1)3 − (n + 1) = (n3 + 3n2 + 3n + 1) − (n + 1)

= 3k + 3n2 + 3n

which is clearly a multiple of 3. Hence Sn+1 is true, and so Sn is true for every n by induction.

20. Look at the first few values: B1 = 1, B2 = 5, B3 = 23, B4 = 119, . . . . If these are compared to the
factorials: 1! = 1, 2! = 4, 3! = 6, 4! = 24, 5! = 120, . . . , it is clear that Bn = (n + 1)! − 1 holds for
n = 1, 2, 3, 4 and 5. So it seems a reasonable conjecture that

Bn = (n + 1)! − 1 for n ≥ 1. (Sn )

This certainly holds for n = 1: B1 = 1 = 2! − 1. If this is true for some n ≥ 1, then

Bn+1 = [1 · 1! + 2 · 2! + · · · + n · n!] + (n + 1)(n + 1)!

= [(n + 1)! − 1] + (n + 1)(n + 1)!
= (n + 1)![1 + (n + 1)] − 1
= (n + 1)![n + 2] − 1
= (n + 2)! − 1

Hence Sn+1 is true and so the induction goes through.

Note that many times mathematical theorems are discovered by “experiment”, somewhat as in this
example. Several examples are worked out, a pattern is observed and formulated, and the result is
proved (often by induction).

22. b. If we know that Sn ⇒ Sn+8 then it is enough to verify that S1 , S2 , S3 , S4 , S5 , S6 , S7 , and S8 are
all true. Then
S1 ⇒ S9 ⇒ S17 ⇒ S25 ⇒ · · ·
S2 ⇒ S10 ⇒ S18 ⇒ S26 ⇒ · · ·
S3 ⇒ S11 ⇒ S19 ⇒ S27 ⇒ · · ·
S4 ⇒ S12 ⇒ S20 ⇒ S28 ⇒ · · ·
S5 ⇒ S13 ⇒ S21 ⇒ S29 ⇒ · · ·
S6 ⇒ S14 ⇒ S22 ⇒ S30 ⇒ · · ·
S7 ⇒ S15 ⇒ S23 ⇒ S31 ⇒ · · ·
S8 ⇒ S16 ⇒ S24 ⇒ S32 ⇒ · · ·
Clearly each Sn will appear in this array, and so will be true.
179
 ADAPTED FORMATIVE ONLINE COURSE COURSE LOGISTICS
OPEN TEXT ASSESSMENT SUPPLEMENTS & SUPPORT

a d v a n c i n g l e a r n i n g

LYRYX.COM

CHAMPIONS OF ACCESS TO KNOWLEDGE