Matrices

Elementary Matrix Theory

It is often desirable to use matrix notation to simplify complex mathematical expressions. The simplifying
matrix notation usually makes the equations much easier to handle and manipulate. Let consider the
following set of n simultaneous algebraic equations:

a x a x  a x  y
11 1 12 2 1n n 1
a x a x  a x y
21 1 22 2 2n n 2

a x a x  a x  yn
n1 1 n2 2 nn n

We may simplify to AX  Y ; A, X, and Y are defined as matrices. These three matrices are defined to be:

 a11 a
12
a
1n
  x1   y1 
     
a a a  x  y 
A   21 22 2n
 X  2  Y  2 
     
a a  x  y 
 n1 nn   n   n 

where the bracketed arrays are simplified of coefficients and variables.

Definition of a Matrix
A matrix is a collection of elements arranged in a rectangular or square array. Several ways of
representing a matrix are as follows:

 0 1  0 1  0 1  
A    a
 2 4   2 4 2 4  ij  2,2

1
Matrix Elements
a a a 
 11 12 13 
A   a 21 a a 
 22 23 
a a a 
 31 32 33 

a is defined as the element, entries, or components in the ith row and the jth column of the matrix
ij
st
(Row 1 and column last).

Column
C C C
1 2 3
  
Row 1  R a a a 
1  11 12 13
Row 2  R a a a 
2  21 22 23 
Row 3  R a a a 
3  31 32 33 

Order of Matrix: refers to the total number of rows and columns of the matrix. In general, a matrix
with n rows and m columns is termed "n x m" or "n by m".

Square Matrix: number of rows is equal to number of columns, when m = n,

Column Matrix: one that has one column and more than one row: mx1 matrix m > 1.

Row Matrix: is one that has one row and more than one column: 1xn matrix n > 1.

Diagonal Matrix: is a square matrix with aij  0 for all i  j

a 0 0 
 11 
 0 a 0 
 22 
 0 0 a 
 33 

Unit Matrix (Identity Matrix): is a diagonal matrix with all the elements on the main diagonal  i  j 
equal to 1. A unity matrix is often designated by either I or U.
1 0 0 0
1 0 0  0 1 0 0
 
Unity example is I  0 1 0  I = 0 0 1 0
 
0 0 1   
0 0 0 1

2
Null Matrix: is one whose elements are equal to zero for example:
0 0 0
O 
0 0 0

Symmetric Matrix: is a square matrix that satisfies the condition aij  a ji . For example:
6 5 1 
   1 4
 5 0 7 ;  4 1 
 1 7 1  
 

Transpose of a Matrix: The transpose of a matrix A is defined as the matrix that is obtained by
interchanging the corresponding rows and columns in A. Let A be an nxm
matrix which is represented by A   a  . Then the transpose of A,
 ij 
n, m

denoted by AT or A is given by AT  transpose of A  a  .

 ij 
m, n
 3 2 1
For example: A   
 0 1 5 
3 0
 
 AT   2 1
1 5 
 

Skew-Symmetric Matrix: is a square matrix that equals its negative transpose, that is:
A   A

3
Summation or Subtraction of Matrix
Given two m x n matrices A  aij  and B  bij  their sum is

A  B   a  b 
 ij ij 

And their difference is

A  B   a  b 
 ij ij 

Example
 5 4   4 8
 3 7    6 0
   
 0 1   5 3
Solution

 5 4  4 8  5  (4) 4  8 
 3 7    6 0   3  6 7  0 
     
 0 1   5 3 0  (5) 1  3 
 9 4 
  9 7 
 5 2 

Example
5 6  4 6 
8 9    8 3
   
Solution
5 6  4 6  5  4 6  6
8 9    8 3  8  8 9  3 
     
 1 0
 
16 6 

4
Scalar Multiplication
The scalar product of a number k and a matrix A is denoted by kA.
 a11 a
12
  ka11 ka
12

kA  k   
 a 21 a   ka
22   21
ka 
22 

If x  n and c  , then cx  n is the vector whose components are c times those of x.

Example
 2 3
Find 5  
0 4 
Solution
 2 3  2(5) 3(5) 
5  
0 4  0(5) 4(5) 
10 15
 
 0 20 

Example
  4 1   1  2
A= 
0
B= 
 3 8 5 
Solution
 4 1   1 2  4(3) 1(3)   1(2) 2(2) 
3 A  2B  3    2  =   
 3 0  8 5   3(3) 0(3)   8(2) 5(2) 
  12 3  2  4  12  2 3  4 
= 
0  16 10   9  16 0  10 
+ =
 9
 14  1
= 
 25 10 

5
Multiplication of Matrices
th th
Let A be an m x n matrix and let B be an n x k matrix. To find the element in the i row and j column
th
of the product matrix AB, multiply each element in the i row of A by the corresponding element in
th
the j column of B, and then add these products. The product matrix AB is an m x k matrix.

Matrix A Matrix B
mxn nxk
Inner must be equal

Outer: Order of AB is m x k

a b   e f   a11 a
12

AB =  .  
c d   g h   a a 
 21 22 

a b   e f   ae  bg _
c d  .  g 
h   _ _ 
a11
  
a b   e f   _ af  bh 
c d  .  g 
h   _ _ 
a12
  

a b   e f  _ _
c d  .  g 
h  ce  dg _ 
a 21
  
a b   e f  _ _ 
c d  .  g 

h   _ cf  dh 
a 22
  

a b   e f   ae  bg af  bh 
. 

c d   g h   ce  dg cf  dh 

C  AB  a b
ij m, n ij n, k

c
ij m, k

k
cij   ain bnj for i  1, 2, ,k and j  1, 2, ,k
n 1

6
Definition
We define the product Ax of a matrix A and the vector x to be the linear combination of the column
vectors of A with coefficients from the vector x

Ax = x1a1 + x2 a2 + + xn an

Example: Given the matrices

b 
a a a   11 
A   11 12 13
 B  b 
 a 21 a
22
a 
23   21 
b31 

 a11b11  a12 b21  a13 b31 

AB   
 a 21b11  a 22 b21  a 23 b31 

Example
1 3 1 0 1 2 
Given: A    B 
7 2  3 1 4 1
Find AB and BA.
Solution

1 3 1 0 1 2 
AB    
7 2  3 1 4 1
1(1)  (3)3 1(0)  (3)1 1(1)  (3)4 1(2)  (3)(1) 

 7(1)  2(3) 7(0)  2(1) 7(1)  2(4) 7(2)  2(1) 

 8 3 13 5 

13 2 1 12 

BA can be found since: B: 2x4 and A: 2x2

Note : AB  BA

7
Determinant of a Matrix: with each square matrix a determinant having the same elements and
order. The determinant of a square matrix A is designated by:

Det A   D  A
A A

a b  a b
Determinant of the matrix   is define as  ad  bc
c d  c d

Example
 3 4
Let A    . Find A
 6 8
Solution

3 4
A
6 8
 3(8)  4(6)
 48

1 0 1
Consider the matrix: A  0 3 2 
1 1 0 

1 0 1
Then: A0 3 2
1 1 0

3 2 0 2 0 3
1 0  (1)
1 0 1 0 1 1
 1(0  2)  0(0  2)  (1)(0  3)

1

Determinant Using Diagonal Method

a a a a a a a a a a a a a a
11 12 13 11 12 11 22 33 12 23 31 13 21 32
a a a a a
21 22 23 21 22
a a a a a a a a a a a a a a
31 32 33 31 32 13 22 31 11 23 32 12 21 33

8
Geometric Interpretation of the Determinant
Let A be the matrix of the vectors v and v
1 2

A  v , v 
 1 2
The vectors v and v determine a parallelogram where the area is equal to A
1 2

In three-dimensions A  v , v , v  , the vectors v , v and v determine a cube where the volume is

 1 2 3 1 2 3
equal to A .

Exercise
x 0 1
Evaluate 2 x x2
3 x 1

Solution
x 0 1 x 0
2 x x2 2 x  x 2  0  2 x   3x   x 4  0
3 x 1 3 x

  x4  x2  5x

9
Singular Matrix: is said to be singular if the value of its determinant is zero. Where a matrix is
singular, it usually means that not all the rows or not all the columns of the matrix
are independent of each other. Let consider the following set of equations:

2 x  3x  x  0
1 2 3
1x  x  x  0
1 2 3
x  2x  2x  0
1 2 3

The third equation is equal to the sum of the first two equations. Therefore these three equations are
not completely independent. In matrix form, these equations may be represented by AX = 0

x 
 2 3 1   1
Where A   1 1 1  & X  x 
 2
 1 2 2   x 3 

Determinant of A:
2 3 1
A  1 1 1  0
1 2 2

 Therefore the matrix is singular.

The n x n matrix A is said to be nonsingular if we can solve the system Ax  b for any choice of the
vector b in n . Otherwise the matrix is singular.

Some Operations of a Matrix Transpose:

 A  A
 kA  kA ; where k is a scalar

 A  B   A  B
 AB   BA

10
Adjoint of a Matrix: Let A be a square matrix of order n. The Adjoint matrix of A denoted by
Adj ( A) is defined as Adj ( A)  ij cofactor of DetAn,n ; where the ij cofactor

of the determinant of A is the determinant obtained by omitting the ith row and
i j
the jth column of A and then multiplying by 1 .

Example
a b 
Determine the Adjoint matrix of: A   
c d 

a b
The determinant of A is A 
c d
 d b 
 adjA   
 c a 

Example

a a a 
 11 12 13 
A   a 21 a a 
 22 23

 a 31 a
32
a 
33 

 a a a a
 22 33 23 32  12
 a a
33
a a
13 32  a a
12 23
a a
13 22


 
Adj  A    a 21 a 33  a 23 a 31  a a
11 33
a a
13 31  11
 a a
23
a a
13 21  

 
 a 21 a 32  a 22 a 31  11
 a a
32
a a
12 31  a a
11 22
a a
12 21 

11
Inverse of a matrix
The n x n matrix A is invertible if there is an n x n matrix B such that AB  I and BA  I .
A matrix B with the property is called an inverse of A .

A is A1 
adjA
A

a b 
A 
c d 
2 by 2 matrices (only)

a b  1  d b
A    A1   
c d  ad  bc c a 

If ad  bc  0 , then A1 doesn’t exist

To find inverse matrix using Gauss-Jordan method:

 A I    I A1  where A1 read as " A inverse "

 

Example
 1 2 1
A    A =?
 3 4 
Solution
4 2
A1 
1
( 1)(4)  ( 2)(3)  3 1

14 2

2  3 1


 2 1 
 3 
  2  1
2

Properties of the Inverse Matrix

 AA1  A1A  I

 
1
 A1 A

  AB 1  B1A1
12
Example
 1 0 2
A   1 2 3  Find A1
 1 1 0 
Solution

 1 0 2 1 0 0 1 2 3 0 1 0 1 1 0 0 0 1
 
 1 2 3 0 1 0  R2  R1 1 0 2 1 0 0 1 0 2 1 0 0
 1 1 0 0 0 1  R  R 0 2 5 1 1 0 0 1 2 1 0 1
3 1

1 0 2 1 0 0 
  1
0 2 5 1 1 0 2 R2 0 1 5 1 1 0
2 2 2
0 1 2 1 0 1 

0 1 2 1 0 1
1 0 2 1 0 0
 0 1 5 1 1 0
5 1 1 0 2 2 2
0 1 2 2 2  1
  0 0 1 1 1
0 1 2 1 0 1 R3  R2 2 2 2

 
1 0 2 1 0 0
0 1 5 1 1 0 0 0 1 1 1 2
 2 2 2 
 1 1 1  2 R3
0 0 2 2 2
1

0 1 5 1 1 0
1 0 2 1 0 0  R1  2 R3 2 2 2
1 0 2 1 0 0
  0 0 5 5  5 5
0 0 2 2 2 4
0 1 52 12 1 0 R 5R
2
 2 2 3 0 1 0
2 2 2
3 2 5 1 0 0 3 2 4
 
0 0 1 1 1 2 

1 0 0 3 2 4 
 
0 1 0 3 2 5
0 0 1 1 1 2 
 3 2 4 
 A1   3 2 5
 1 1 2 

13
 -1
Solving a System Using A

To solve the matrix equation AX = B.

o X: matrix of the variables
o A: Coefficient matrix
o B: Constant matrix

AX  B
A1  AX   A1B Multiply both side by A
-1

 A1A X  A1B Associate property

IX  A1B Multiplicative inverse property

X  A1B Identity property

Example

Solve the system using A1

x  2z  6  3  2  4
1  
 x  2 y  3z  5 Given A  3  2  5
 
x y 6  1 1 2 
Solution
 1 0 2  x  6
 1 2 3   y    5
     
 1 1 0   z   6 
A X  B

X  A1B
 x   3 2 4   6   3(6)  2( 5)  4(6)  18  10  24 
 y    3 2 5 .  5   3(6)  2(5) 5(6)   18  10  30 
 
       
 z   1 1 2   6   1(6) 1(5)  2(6)   6  5  12 

4
  2 
 
 1 

Solution: {(4, -2, 1)}

14
Cramer's Rule
Given:
a xb y  c
1 1 1
a xb y  c
2 2 2

D Dy
If D  0 x x y
D D

a b c b a c
1 1 1 1 1 1
D D  D 
a b x c b y a c
2 2 2 2 2 2

Example
Use Cramer's rule to solve the system
5 x  7 y  1
6x  8 y  1
Solution
5 7
D  2
6 8

1 7
D   15
x 1 8

5 1
D   11
y 6 1
D
x x  15  15
D 2 2
D
y
y  11   11
D 2 2

Solution: 15 ,  11
2 2  

15
 b1 a12 a13 b1 a12
D x  b2 a 22 a 23 b2 a 22
b3 a32 a33 b3 a32
D b a a a a b a b a a a b b a a a b a
x 1 22 33 12 23 3 13 2 32 13 22 3 1 23 32 12 2 33

a11 b1 a13 a11 b1

Dy  a 21 b2 a 23 a 21 b2
a31 b3 a33 a31 b3

a11 a12 b1 a11 a12

D  a 21 a 22 b2 a 21 a 22
z
a31 a32 b3 a31 a32

D Dy D
x x y z z
D D D

Example
Use Cramer's rule to solve the system
x – 3y + 7z = 13
x+ y+ z=1
x – 2y + 3z = 4
Solution
1 3 7 13 3 7
D1 1 1  10 Dx  1 1 1  20
1 2 3 4 2 3

1 13 7 1 3 13
Dy  1 1 1  6 Dz  1 1 1  24
1 4 3 1 2 4

20
x  2
10
6
y  0.6
10
24 12
z 
10 5

Solution:  2, 0.6, 125 

16
Properties of Matrix
Addition and Scalar Multiplication

A B  B  A Commutative Property of Addition

A  ( B  C )  ( A  B)  C Associative Property of Addition

(kl ) A  k (lA) Associative Property of Scalar Multiplication

k ( A  B)  kA  kB Distributive Property

(k  l ) A  kA  lA Distributive Property

A 0  0 A  A Additive Identity Property

A  ( A)  ( A)  A  0 Additive Inverse Property

Multiplication

A( BC )  ( AB)C Associative Property of Multiplication

A( B  C )  AB  AC Distributive Property

( B  C ) A  BA  CA Distributive Property

17