SIF1001/sMES1103/KHChew/12092017/V1

Lecture Notes
SMES1103 BEGINNING MATHEMATICAL METHODS
SIF1001 MATHEMATICAL METHODS I

LINEAR ALGEBRA

Main Textbook
Chapter 3, Mathematical methods in the physical sciences (3rd ed) by Mary L. Boas

Contents
• Introduction
• Matrix and Row Reduction
o transpose of matrix; sets linear equations; augmented matrix; Determinants;
Cramer’s rule
• Vectors
o Notation; magnitude; addition of vectors; multiplication by a constant &
substration; unit vector; vector in tems of components; multiplication of vector;
perpendicular and parallel vectors
• Matrix operations
o Matrix equations; multiplication of a matrix by a number; addition of matrices;
multiplication of matrices; zero matrix; identity matrix or unit matrix; operation
with determinants; applications of matrix multiplication; inverse of matrix
• Linear combinations, Linear functions, linear operators; Matrix operators, linear
transformations; eigenvalues and eigenvectors; Special matrices

1
SIF1001/sMES1103/KHChew/12092017/V1

Introduction
• Problems in various fields of science and mathematics involve the solution of sets of linear
equations. It is algebra, but, it has a useful geometric interpretation
• Vectors are defined to have both a magnitude and a direction. A scalar (such as speed) just has a
magnitude given by its value.

Matrix and Row Reduction

• A matrix is a rectangular array of quantities, usually enclosed in large parentheses

 1 5 2 
A   2  3 (2 by 3) matrix
 3 0 6 
Aij : i  row number, j  column number
A11  1, A12  5, A13  2, A21  3, A22  0, A23  6.

• Transpose of a matrix
 1 3 
 
A   5 0   3  2 matrix
T

 2 6 
 
A  T
ij
 A ji

• Sets of Linear Equations

2 x  z  2

6 x  5 y  3z  7
2 x  y  4


 2 0 1  x   2 
    
 Mr  k,  6 5 3   y    7  ,
 2 1 0   z   4 
    
 2 0 1 x  2
     
where M   6 5 3  , r   y , k   7 .
 2 1 0  z  4
     

M x
j 1
ij j  ki , i  1,2,3.

2
SIF1001/sMES1103/KHChew/12092017/V1

• Augmented matrix

2 x  z  2  2 0 1 2 
  
6 x  5 y  3z  7 A  6 5 3 7
2 x  y  4  2 1 0 4 
  

(a) Eliminate the x terms in the other two equations by using the first equation.
e.g. eq.(2) – eq.(1) x 3  eq.(2) , eq.(3) – eq.(1)  eq.(3)
2 x  z  2  2 0 1 2 
  
 5y  6z  1  0 5 6 1
  y   2  
 z  0 1 1 2 

(b) For convenience, interchange the second and third equations

2 x  z  2  2 0 1 2 
  
  y  z  2  0 1 1 2 
  6z  1  
 5y  0 5 6 1

(c) Eliminate the y terms by using the second equation.

e.g. eq.(3) – eq.(2) x 5  eq.(3)
2 x  z  2  2 0 1 2 
  
  y  z  2  0 1 1 2 
 11z  11  
  0 0 11 11

(d) Eliminate the z terms by using the third equation.

e.g. eq.(1) + eq.(3) / 11  eq.(1) , eq.(2) – eq.(3) / 11  eq.(2)
2 x  3 2 0 0 3 
  
  y  1  0 1 0 1 
 11z  11  0 0 11 11
  

(e) finalizing
2 x  3  1 0 0 3 / 2
  
 y  1  0 1 0 1 
 z   1 
 1 0 0 1

Allowed rules:
i. Interchange two rows
ii. Multiply (or divide) a row by a (nonzero) constant
iii. Add a multiple of one row to another; this includes subtracting, that is, using a negative
multiple.

3
SIF1001/sMES1103/KHChew/12092017/V1

Example 2.

x  y  4z  5

2 x  3 y  8z  4
 x  2y  4z  9


 1 1 4 5   1 1 4 5   1 1 4 5 
  with the 1st row   with the 2nd row  
 2 3 8 4     0 1 0 6     0 1 0 6 
 1 2 4 9   0 1 0 4   0 0 0 10 
     

We cannot get an answer the equations are inconsistent.

Determinants
• Recall that a matrix is simply a display of a set of numbers; it does not have numerical value.
o For a square matrix, however, there is a useful number called the determinant of the
matrix.

• Evaluating determinants:
(i) 2 by 2 matrix
a b  a b
A , det A   ad  bc.
c d  c d

(ii) nth order matrix

a11 a12 a13 a1n
a21 a22 a23 a2 n
a31 a32 a33 a3 n

an1 an 2 an 3 ann
When removing the row and the column containing the element a ij, we have the remaining
determinant, Mij, called a minor of aij. For example

a11 a12 a13 a1n

a21 a22 a23 a2 n
M33  a31 a32 a33 a3 n

an1 an 2 an 3 ann

 (n-1) by (n-1) determinant

4
SIF1001/sMES1103/KHChew/12092017/V1

Sign

   
   
    etc
   
etc
 
 

Cofactor:

 1
i j
 Mij

Finally, multiply each element of one row (or one column) by its cofactor and add the results.

  1
i j
det A  aij  M ij
i ( or j )

Example 1.

1 5 2
1 5
7 3 4  a23  4, M23 
2 1
2 1 5

(i) Approach 1 (third column)

1 5 2
7 3 1 5 1 5
7 3 4 2 4 5  2  1  4  11  5  38  148.
2 1 2 1 7 3
2 1 5

(ii) Approach 2 (first row)

1 5 2
3 4 7 4 7 3
7 3 4 1 5 2  11  135  2  148.
1 5 2 5 2 1
2 1 5

5
SIF1001/sMES1103/KHChew/12092017/V1

Example 4. Evaluate the determinant

4 3 0 1
9 7 2 3
D
4 0 2 1
3 1 4 0

0 3 2 1
0 3 2 0 3 2
3 7 4 3
  1 3 7 4   1 3 7 4
0 0 0 1
3 1 4 0 6 0
3 1 4 0
0 2
  1   1  6   36
3 4

1. Subtract 4 times the fourth column from the first column, and subtract 2 times the fourth column
from the third column

2. Do a Laplace development using the third row

3. Add the second row to the third row.

4. Do a Laplace development using the third row.

Useful facts about determinants:

1. If each element of one row (or one column) of a determinant is multiplied by a number k, the
value of the determinant is multiplied by k.

2. The value of a determinant is zero if,

(a) all elements of one row are zero

(b) two rows ( or two columns) are identical

(c) two rows (or two columns) are proportional.

3. If two rows (or two columns) of a determinant are interchanged, the value of the determinant
changes sign.

4. The value of a determinant is unchanged if

(a) rows are written as columns are columns as rows

(b) we add to each element of one row, k times the corresponding element of another row, where k
is any number (and a similar statement for columns).

6
SIF1001/sMES1103/KHChew/12092017/V1

Cramer’s rule

a1x  b1y  c1 a b1   x   c1 
  1        Mr  k
a2 x  b2 y  c2  a2 b2   y   c2 

solving the set of linear equations give

c1b2  c2 b1 a c  a2c1
x ,y 1 2
a1b2  a2b1 a1b2  a2b1
or
c1 b1 a1 c1
c2 b2 1 c1 b1 a2 c2 1 a1 c1
x  , y  .
a1 b1 D c2 b2 a1 b1 D a2 c2
a2 b2 a2 b2
a1 b1
where D 
a2 b2

Denominator: determinant of the matrix with coefficients in the left side (M)

- Numerator: for x, replace x part in M with right side part, and take the determinant.

for y, replace y part in M with right side part, and take the determinant.

We can use this method when you get the solution for the n linear equations (in case that D  0).

Example 5. Solve the set of equations

2 x  3 y  3,

 x  2y  5.
We have

2 x  3 y  3,  2 3  x   3 
      
 x  2y  5.  1 2  y   5 
2 3
D  4  3  7
1 2
1 3 3 6  15 1 2 3 10  3
x   3, y    1.
D5 2 7 D 1 5 7

7
SIF1001/sMES1103/KHChew/12092017/V1

Vectors
• Notation

A  A   Ax , Ay 

• Magnitude

A  A  Ax2  Ay2  Az2

Example 1. If a force F has an x component of 4 N and a y

component of 3 N. Determine the magnitude of |F|.

Fx  4 N, Fy  3 N

3
Then, F | F | 42  32  5 and   arctan
4

• Addition of vectors

A B  B A : commutative law for addition

 A  B   C  A  B  C  : associative law for addtion

• Multiplication by a constant & subtraction

8
SIF1001/sMES1103/KHChew/12092017/V1

• Unit vector
A
A

• Vectors in terms of components

A  iAx  jAy  kAz

A  B  i  Ax  Bx   j  Ay  By   k  Az  Bz 

• Multiplication of vectors 1: Scalar Product (or dot product)

A  B  A B cos
A B  B  A

A  B  A   projection of B on A  A cos
 B  projection of A on B 

A  A  A2
B cos
A  B  C   A  B  A  C

A  B  Ax Bx  Ay By  AzBz

• Angles between two vectors using scalar product

A B
cos  
A B

9
SIF1001/sMES1103/KHChew/12092017/V1

Example 3. Find the angles between these two vectors

A  ( Ax , Ay , Az )   3,6,9   3i  6 j  9k, B   2,3,1  2i  3 j  k

A  B  3   2   6  3  9  1  21
A  32  62  92  3 14, B  14
A B 21 1
cos     ,   60
A B 3 14  14 2

• Perpendicular and Parallel vectors

A B  A  B  Ax Bx  Ay By  Az Bz  0
Ax Ay Az
A / /B  A  cB   
Bx By Bz

• Multiplication of vectors 2 : Vector Product (or cross product)

A  B : vector product (right-handed rule)

A  B  A B sin ,
where A  B  A, B and  is positive  180

A  B  0  A / /B
AA  0
A  B  B  A

ii  j j  k k  0
i  j  k, j  k  i, k  i  j, j  i  k.

10
SIF1001/sMES1103/KHChew/12092017/V1

i j k
A  B  iAx  jAy  kAz   iBx  jBy  kBz   Ax Ay Az
Bx By Bz

Example 4. Find a vector perpendicular to both A and B, where

A   2,1, 1 , B  1,3, 2 
Solution:
i j k
A  B  2 1 1  i  3 j  5k.
1 3 2

Matrix operations

• Matrix equations

 x r u 2 1 5 
  
 y s v   3 7i 1  i 
x  2, y  3, r  1, s  7i , u  5, v  1  i .

• Multiplication of a matrix by a number

 2
A  2i  3 j  A    or AT   2 3 
3

a c e   ka kc ke 
k  
b d f   kb kd kf 

• Addition of Matrices

 1 3 2   2 1 4   1  2 3  1 2  4   3 2 2 
    
 4 7 1   3 7 2   4  3 7  7 1  2   7 0 1

 1 3 2   2 1
note:   ?
 4 7 1  3 5 

11
SIF1001/sMES1103/KHChew/12092017/V1

• Multiplication of matrices
o The element in row i and column j of the product matrix AB is equal to row i of A times
column j of B. [(number of row i of A) = (number of column j of B)] .
o In index notation,

 AB ij   Aik Bkj

k

For example,
 a c  e f   ae  cg af  ch 
AB     C
 b d  g h   be  dg bf  dh 

Example 1.
 4 2 1 5 3 
A , B   
 3 1   2 7 4 
 4 2  1 5 3   4  1  2  2 4  5  2  7 4  3  2  4   8 34 4 
AB      
 3 1  2 7 4   3  1  1 2 3  5  1 7 3  3  1 4   1 8 13 
Example 2. Find AB and BA

 3 1  5 2
A , B   
 4 2   7 3 
 22 3   7 1
AB    , BA    “not commutative”
 34 2   33 13 

• Zero matrix
o zero or null matrix means one with all its elements equal to zero.

NOTE:
 2 4  0 0
M   M  but, M  0
2

 1 2  0 0

• Identity matrix or Unit matrix

 1 0 0
 
I   0 1 0 , IA  AI  A
 0 0 1
 

• Operation with Determinants

det AB  det BA  det A  det B

• Applications of matrix multiplication

12
SIF1001/sMES1103/KHChew/12092017/V1

 1 0 1  x   5 
    
 2 3 0   y    1 
 1 3 2   z   10 
    

 xz   5 
   
Method 1  2 x  3 y    1 
 x  3 y  2z   10 
   
 1 0 1  x   5 
     1
Method 2  2 3 0   y    1   Mr  k, Then r  M k
 1 3 2   z   10 
    

• Inverse of a Matrix

MM1  M1M  I
1
where Cij  cofactor of mij ,  1
i j
M1  CT  Mij
det M

Example 3. Find M-1

 1 0 1
 
M   2 3 0  , det M  3
 1 3 2 
 

Finding the cofactor

3 0 2 0 2 3
1st row :  6,   4,  3,
3 2 1 2 1 3
0 1 1 1 1 0
2nd row :   3,  3,   3,
3 2 1 2 1 3
0 1 1 1 1 0
3rd row :  3,   2,  3.
3 0 2 0 2 3

6 4 3 6 3 3
  1 1  
C   3 3 3  so M-1  CT   4 3 2 
3 2 3 det M 3 
  3 3 3

13
SIF1001/sMES1103/KHChew/12092017/V1

Linear combinations, linear functions, linear operators

• A function of a vector, say f(r), is called linear if

f r1  r2   f r1   f r2  , and f  ar   af r 

• F(r) is a linear vector function if

F r1  r2   F r1   F r2  , and F  ar   aF r 

• Linear operator

O  A  B   O  A   O B  and O  kA   kO  A 

Matrix operators, Linear transformations

• One set coordinate & r  R

Moving a point to some other point  mapping or transformation

 X  ax  by ,  X   a b  x 
 or     , or R  Mr
Y  cx  dy , Y   c d  y 
M : transformation matrix (linear operator)

Eigenvalues and eigenvectors

• Definition of eigenvalue and eigenvector

Some vectors are not changed in direction by transformation

R  Mr  R  r where   const. r : eigenvectors (characteristic vector)

: eigenvalues

14
SIF1001/sMES1103/KHChew/12092017/V1

Eigenvalues

According to the definition, consider a transformation

 X   5 2  x 
    .
 Y   2 2  y 

 X   5 2  x   x   x 
           Eigenvector condition: R  r 
 Y   2 2  y   y   y 
 5 x - 2y  μx,  (5   ) x - 2y  0,
 or 
 2 x  2y   y ,  2 x  (2   )y  0.

Condition for a solution other than x  y  0,

5 2
0 characteristic equation of matrix M.
2 2
(5   )(2   )  4   2  7   6  0,
  1 or   6.  eigenvalues 

Eigenvectors

From the above results

 2x  y  0 for   1

 x  2y  0 for   6.

 R  r on 2 x  y  0  5 2 
 through  .
 R  6r on x  2y  0  2 2 
Any points in two straight lines can be eigenvectors

Example

 5 2  1   5-4   1 
         for   1
 2 2  2   -2  4   2 
 5 2  2   10  2   12  2 
          6   for   6
 2 2  -1  -4-2   6   1

15
SIF1001/sMES1103/KHChew/12092017/V1

Special matrices
Below are list of special matrices, their notation and names.

16