1a. Matrix Algebra

Mathematics II – MATH 1211
UNIT 1 MATRICES
Unit Structure
1.0 Overview
1.1 Learning Objectives
1.2 List of Special Symbols
1.3 Basic Concepts and Definitions
1.4 Determinant of a Matrix
1.5 Cofactor Method for A-1
1.6 Properties of Determinants
1.7 Summary
1.8 Supplementary Exercises
1.9 Answers to Activities & Supplementary Exercises
1.0 OVERVIEW
Matrices and determinants constitute the structure of matrix algebra that pervades and
enriches almost all areas in science and engineering computations. In this Unit, various
types of matrices are defined before introducing briefly the topic on the algebra of matrices
in general. The determinant of a matrix is then treated in detail, including the main
properties associated with it. The co-factor method for finding the inverse of a matrix is
also introduced.
Unit 1 1
1.1 LEARNING OBJECTIVES
By the end of this Unit, you should be able to do the following:
1. Do matrix operations
2. Compute the determinant of a matrix
3. Understand the properties of the determinant of a matrix
4. Determine the inverse of a matrix using the co-factor method.
1.2 LIST OF SPECIAL SYMBOLS
A a general matrix
aij (i, j) entry in matrix A
aij complex conjugate of aij
 
A  a ij
mn
matrix A of order m by n
Aij cofactor of aij
M ij minor of aij
D determinant of a matrix
A determinant of A
Adj A adjoint matrix A
A 1 inverse of matrix A
x column vector
xi ith component of vector x
Unit 1 2
ai ith row vector of A

aj jth column vector of A
Ri ith row of matrix
Cj jth column of matrix
1.3 BASIC CONCEPTS AND DEFINITIONS
Matrix
A matrix defines a rectangular array of numbers or symbols. An m x n matrix A has m

rows and n columns. aij = (i, j) entry represents the number in the ith row, jth column.
 a11 a12  a1n   a11 a12  a1n 

a  
a 22  a 2 n   a 21 a 22  a2n 
A  or
21
. (1)
           
   
a m1 am 2  a mn   a m1 a m2  a mn 
Square Matrix
If the number of rows of a matrix equals the number of columns ( m = n), then the matrix
is called a square matrix of order n. The principal or, leading diagonal of the square
matrix defines the entries a11 , a12 ,..., a nn . For example, the matrices
b11 b12 b13 

a11 a12   
A  and B  b21 b22 b23 ,
a 21 a 22  b31 b32 b33 
Unit 1 3
are square matrices of order two and three respectively. The elements a11 , a 22 and
b11 , b22 , b33 constitute the principal diagonals of the matrix A and B respectively.
Diagonal Matrix
A square matrix A of the form
 a11 0  0 
0 a 22  0 
A 
     
 
0 0  a nn 
is called a diagonal matrix. Here, A  aij   nxn

with aij  0, i  j, i, j = 1, ..., n .
Zero Matrix, 0
The zero matrix, 0 is a matrix with all entries zero.
Unit or Identity Matrix
The identity or the unit matrix is a diagonal matrix in which all the diagonal elements
aii , i = 1, ..., n are equal to 1. It is denoted by I and is written as
1 0  0
0 1  0
I .
    
 
0 0  1
Unit 1 4
Upper Triangular Matrix
A square matrix U of the form
a11 a12  a1n 

0 a 22  a2n 
U ,
     
 
0 0  a nn 
is called an upper triangular matrix. Note that all the entries below the main diagonal are
zero.
Lower Triangular Matrix
A square matrix L of the form
 a11 0  0 
a a 22  0 
L ,
21
     
 
a n1 an 2  a nn 
is called a lower triangular matrix. This time, all the entries above the main diagonal are
zero.
Unit 1 5
Transpose of a Matrix
 
If A  aij
mxn
T
, then the transpose of A, denoted by A , is obtained by interchanging the
rows and columns of the matrix A, that is, AT  a ji  nxm

. Symbolically,
 a11 a 21  a m1 
a a 22  am2 
A  .
T 12
     
 
a1n a2n  a mn 
T
Clearly  A T   A . For example, if
 1 5 7 6
A   2 3 4 5,
 0 0 1 0
then,
1  2 0
5 3 0
AT   .
7 4 1
 
6 5 0
Unit 1 6
Symmetric Matrix
If AT  A, the matrix A  aij   nxn

is called symmetric. Examples of symmetric matrices
are
2 3 1 9
1 4 0  
1 5  3 4 0 2
 ,  4 2  7 ,  , etc.
 5 7 0  7  3  1 0 5 8
 9 
 2 8 1 
Skew-Symmetric Matrix
 
A square matrix A  aij
nxn
is called skew-symmetric if A T   A . Clearly the diagonal
elements are zeros because aii  aii if aii  0 for all i. For example
0 1 2
A   1 0 3,

 
 2  3 0
is a skew-symmetric matrix.
Hermitian Matrix
 
A square matrix A  aij
nxn
is called a Hermitian matrix if
AT  A ,
Unit 1 7
 
where A  aij , and the overbar denotes complex conjugate. Hence
a ji  aij , i, j = 1, ..., n
for an n x n Hermitian matrix A. It follows that the elements along the principal diagonal
of a Hermitian matrix are always real. For example, one can easily verify that
 2 1  i 0 

A   1  i 2  i ,
 0 i 2 
is a Hermitian matrix.
If, in addition, A is a real Hermitian matrix, then A T  A , that is, A is a symmetric matrix.
Skew-Hermitian Matrix
A square matrix A  aij   nxn

is called a skew-hermitian matrix if AT   A.
Hence, a ji   aij , i, j  1, ..., n. . It follows that the elements along the principal diagonal of
a skew-hermitian matrix are either zeros or pure imaginaries. For example,
 0 2i 1

A  2  i i i ,
  1 i 2i 
a Skew-Hermitian matrix.
Unit 1 8
If, in addition, A is a real Skew-Hermitian, then A T   A , that is, A is a skew-symmetric
matrix.
Note: In general AB ≠ BA, that is, matrix multiplication is not commutative.
1 1 1  2 0 4 4 
     
(1) A   2 0 1 , B   1 1 . BA is not defined although AB   5 3 .
1 2 4  1 3  8 14 
     
 2 0
 2 1 3  
(2) A   , B   1 1  . AB is 2  2 whereas BA is 3  3 .
1 2 1  1 3
 
 2  1  2 0
(3) A   , B   . AB and BA are of the same size. However
1 2   1 1
 3  1  4  2
   AB  BA   .
4 2  3 1 
MATRIX RULES OF ALGEBRA
Addition is commutative
AB  B A
Unit 1 9
Associative Laws
( A  B)  C  A  ( B  C )
( AB)C  A( BC )
 ( AB)  (A) B  A(B)
Distributive Laws
A( B  C )  AB  AC
( A  B)C  AC  BC
 ( A  B)  A  B
Additive Identity
A 0  A
A A  0
0 A  A0  0
Multiplicative Identity
AI  IA  A
Unit 1 10
1.4 DETERMINANT OF A MATRIX
To each matrix A  aij  nxn

is associated a unique real number called the determinant of
A. This number contains important information about the matrix and is written as
a11 a12  a1n

a12 a 22  a 2n
A . (2)
   
a n1 a n 2  a nn
Sometimes the capital letter D is used to denote A .
Note: The determinant of a rectangular m x n matrix (m  n) is not defined.
Definition
For 2 x 2 matrices:
a b  a b
A   , A   ad  bc.
c d  c d
To extend this to an n x n matrix the following two definitions are required.
Definition
The (i, j) minor of A is the determinant of the (n - 1) x (n - 1) matrix obtained by

removing the ith row and jth column of A, denoted Mij.
Unit 1 11
Example 1
 1 2  3
  4 1 2 3
A   4  1 1 , M 13   13 and M 21   11.
1 3  1 3 3 1
 1 
Definition
The (i,j) cofactor of A is
Aij = (-1)i+j M ij . (3)
From Example 1, A13  (1) (13) M 13  13 and A21  (1) ( 21) M 21  11 .
It follows that the cofactor Aij is either Mij or - Mij depending on whether the sum i + j is
even or odd. Aij can be associated with the entry aij as follows.
1) Cross out the row and the column containing aij.
2) Evaluate the determinant of the (n - 1) x (n - 1) matrix which remains.
3) Assign a "+" or "-" sign according to the position of aij on the checkerboard below,
        
   
     and in general       .
        
   
Unit 1 12
Development of a Determinant
If A is an n x n matrix, then the determinant of A can be computed by multiplying the

entries of any row (or column) by their cofactors and summing the resulting products. If
 a11 a12  a1n 

a a 22  a 2n 
A ,
21
     
 
 a m1 am2  a mn 
A  a11 A11  a12 A12    a1n A1n (cofactor expansion by row 1).
A  a 21 A21  a 22 A22    a 2 n A2 n (cofactor expansion by row 2).
A  a i1 Ai1  a i 2 Ai 2    ain Ain (cofactor expansion by row i).
A  a1 j A1 j  a 2 j A2 j    a nj Anj (cofactor expansion by column j).
Example 2
1 2 3
 
Evaluate the determinant of the matrix A   0 4 0 .
5 6 7
 
Solution
The cofactor expansion by the first row is given by
4 0 0 0 0 4
A  1C11  2C12  3C13  1M 11 - 2M 12  3M 13  1 2 3  32.
6 7 5 7 5 6
Unit 1 13
Similarly the cofactor expansion by the second row is given by
1 3
A  0C 21  4C 22  0C 23  4 M 22  4  4(7  15)  32.
5 7
As an exercise, you can check A using the cofactor expansion by column one.
Note: It is better to choose to expand the determinant using a row or column which will
lead to less or simpler computations. For instance, it is easier and quicker to
expand by the row or column that contains zero entries.
Example 3
2  x 2 3
Solve the equation 2 1 x  6  0.
1 2 x
Solution
The development of the determinant by the first column may be written as
1 x  6 2 3 2 3
 2  x  2  0
2  x  2  x 1 x  6
 2  x  x1  x   12  2 2 x  6   12  31  x   0
or  x 3  x 2  21x  45  0
It is easily verified that x = -3, -3 and 5 satisfy the above equation.
Unit 1 14
Activity 1
1. Evaluate the determinants:

1 2
(i) ,
3 4
1 3 3
(ii) 2 6 3,
1 1 2
2 1 1
(iii)  1 10 1 .
2 3 1
4x  3 2 1
2. Solve the equation x 3 4  0.
2x 1 1 1
1.5 COFACTOR METHOD FOR A-1
If A  0, we show that A 1 exists by constructing it as follows:
First we need to define the adjoint matrix.
Unit 1 15
Adjoint Matrix
The matrix of cofactors of A defines the matrix whose (i,j) entry is Aij, the (i,j) cofactor of
A. The transpose of the matrix of the cofactors is called the adjoint matrix and is
 A11 A21 An1 

A A22 An 2 
Adj  A   
12
,
    
 
 A1n A2n Ann 
denoted by Adj A .
1 2  1
 
For example, if A  3 6 1
3 3 2 
then
A11  9, A12   3, A13   9,

  7, A22  5, A23  3,
A21
 8, A32   4, A33  0,
A31
and
 9 7 8 
Adj  A    3 5  4 .
 9 3 0 
Unit 1 16
Definition
 
A  aij
nxn
is invertible or non-singular if and only if there exists a matrix A 1 such that
AA 1  A 1 A  I . A 1 is given by
1
A1  Adj A. (4)
A
From (4) it follows that A 1 exists if and only if A  0.
If A 1 exists, that is A is invertible, then A is called a non-singular matrix. If A 1 does

not exist then A is called a singular matrix.
Properties of Inverse
 A 1 is unique.
 ( A 1 ) 1  A.
1 1
 (A) 1  A ,  a scalar.

1
 A 1  .
A
 If A 1 , B 1 exist, then ( AB) 1  B 1 A 1 .
Note: There is no way to define matrix division. Only invertible matrices have a
multiplicative inverse. But be careful! Matrix multiplication is not commutative; if
you multiply an equation through by a matrix, you must multiply both sides on the
right or both sides on the left.
Unit 1 17
Example 4
If possible, determine the inverse of the following matrices:
1 2  1 0 2 3 a11 0 0
a11 a12      
A , B  3 6 1 , C   - 2 0 4 , D   0 a 22 0 .
a a
 21 22  3 3 2   3  4 0  0 0 a33 
Solution
1  a 22  a12 
A 1  .
a11 a 22  a 21 a12   a 21 a11 
 9  7 8   9 / 12  7 / 12 8 / 12 
1 1 
B 1
 Adj B     3 5  4    3 / 12 5 / 12  4 / 12.
B 12
  9 3 0   9 / 12 3 / 12 0 
C-1 does not exist because C  0. In fact, C is a singular matrix.
1 / a11 0 0 
D 1 
 0 1 / a 22 0  .
 0 0 1 / a33 
Unit 1 18
Activity 2
For each of the matrices given below, compute the adjoint matrix and hence determine the
corresponding inverse matrix.
1 0 1
(i) A  2 1 1 ,
 
1 1 2
1 0  1
(ii) B  2 2 3 .
 
1 1 0 
1.6 PROPERTIES OF DETERMINANTS
1. If a row (or column) of A consists entirely of zeros, then A  0.
2. AT  A .
3. If A  I , then A  1.
4. If any two rows (or columns) of A are interchanged (or swapped) producing a
matrix B , then B   A .
5. If two rows (or columns) of A are identical, then A  0.
Unit 1 19
6. If any row (or column) of A is multiplied by a scalar k producing a matrix B ,

then B  k A . For example,
a11 ka12 a13 a11 a12 a13 ka11 ka12 ka13

a 21 ka 22 a 23  k a 21 a 22 a 23  a 21 a 22 a 23 .
a31 ka 32 a33 a31 a32 a 33 a31 a 32 a 33
7. If a scalar multiple of one row (or column) is added to some other row (or
column) producing a matrix B then B  A . For example, if
a a12  a 
 11  11 a12
A   , then
 and B   
a 21

a 22 
 a 21  ka11 a 22  ka 12 
a11 a12 a11 a12 a11 a12

B    A k
a 21 a 22 ka11 ka12 a11 a12
 A  0 ( using property 5.)

 A.
8. If A1 , A2 , A3 , ... are all n x n matrices then A1 A2  A1 A2 and, in general,
A1 A2 A3 ...  A1 A2 A3 ...
9. If A is upper triangular, lower triangular, or diagonal, the determinant of A is given

by the product of the diagonal entries.
Unit 1 20
10. If the cofactors of one row are multiplied by the entries of a different row, the
result is zero.
Example 5
1 a bc
Show that 1 b c  a  0 .
1 c a b
Solution
1 a bc 1 a a bc
1 b c  a  1 b a  b  c (using property 7.)
1 c a b 1 c a bc
1 a 1
 (a  b  c) 1 b 1 (using property 6.)
1 c 1
 0 (using property 5.).
Activity 3
By using the properties of the determinant, show that
xa a a
(i) a xa a  x 2 ( x  3a).
a a xa
Unit 1 21
x2 x 1
(ii) y 2
y 1   x  y  x  z  y  z .
z2 z 1
1.7 SUMMARY
In this Unit, you have studied matrices and their operations, including the determinant of a
matrix. You have learnt the different properties of the determinant of a matrix, as well as,
applying the co-factor method to obtain the inverse of a square non-singular matrix.
1.8 SUPPLEMENTARY EXERCISES
1. Given that a  0, show that for all real x
e ax ae ax
(i)  0,
xe ax 1  ax e ax
e ax sin x e ax cos x
(ii)  0,
e ax (cos x  a sin x) e ax (a cos x  sin x)
2. Solve the following determinantal equations:

x x x
(i) y x x  0, x  0.
0 y x
x3 5 1
2
(ii) 8 x  1 1  0.
1 x2  4 1
Unit 1 22
3. By using the properties of the determinant, show that
x1  y1 x2  y2 x3  y3 x1 x2 x3
y1  z1 y2  z2 y 3  z 3  2 y1 y2 y3 .
z1  x1 z2  x2 z3  x3 z1 z2 z3
4. Prove that
f g h f ' g ' h' f g h f g h

d
p q r  p q r  p' q ' r '  p q r .
dx
u v w u v w u v w u ' v ' w'
1.8 ANSWERS TO ACTIVITIES & SUPPLEMENTARY

EXERCISES
Activity 1
1. (i) -2; (ii) 12; (iii) 46.
26
2.  .
15
Activity 2
 1 1  1
1
(i) A 1
 3 1 1 .
2 
 1  1 1
Unit 1 23
3 1  2
1
(ii) B 1
  3  1 5 . .
3
 0 1  2
Supplementary Exercises
2. (i) x y.
(ii) x  1,2,2 / 3.
Unit 1 24

1a. Matrix Algebra

Uploaded by

Copyright:

Available Formats

1a. Matrix Algebra

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1a. Matrix Algebra

Uploaded by

Copyright:

Available Formats

Mathematics II – MATH 1211

1.1 LEARNING OBJECTIVES

By the end of this Unit, you should be able to do the following:

2. Compute the determinant of a matrix

3. Understand the properties of the determinant of a matrix

4. Determine the inverse of a matrix using the co-factor method.

1.2 LIST OF SPECIAL SYMBOLS

aij complex conjugate of aij

Aij cofactor of aij

Adj A adjoint matrix A

ai ith row vector of A

1.3 BASIC CONCEPTS AND DEFINITIONS

A matrix defines a rectangular array of numbers or symbols. An m x n matrix A has m

 a11 a12  a1n   a11 a12  a1n 

b11 b12 b13 

A square matrix A of the form

is called a diagonal matrix. Here, A  aij   nxn

The zero matrix, 0 is a matrix with all entries zero.

Unit or Identity Matrix

Upper Triangular Matrix

A square matrix U of the form

a11 a12  a1n 

Lower Triangular Matrix

A square matrix L of the form

rows and columns of the matrix A, that is, AT  a ji  nxm

If AT  A, the matrix A  aij   nxn

A square matrix A  aij   nxn

a skew-hermitian matrix are either zeros or pure imaginaries. For example,

If, in addition, A is a real Skew-Hermitian, then A T   A , that is, A is a skew-symmetric

Note: In general AB ≠ BA, that is, matrix multiplication is not commutative.

MATRIX RULES OF ALGEBRA

 ( AB)  (A) B  A(B)

1.4 DETERMINANT OF A MATRIX

To each matrix A  aij  nxn

a11 a12  a1n

Sometimes the capital letter D is used to denote A .

Note: The determinant of a rectangular m x n matrix (m  n) is not defined.

To extend this to an n x n matrix the following two definitions are required.

The (i, j) minor of A is the determinant of the (n - 1) x (n - 1) matrix obtained by

The (i,j) cofactor of A is

Aij = (-1)i+j M ij . (3)

1) Cross out the row and the column containing aij.

2) Evaluate the determinant of the (n - 1) x (n - 1) matrix which remains.

If A is an n x n matrix, then the determinant of A can be computed by multiplying the

 a11 a12  a1n 

A  a 21 A21  a 22 A22    a 2 n A2 n (cofactor expansion by row 2).

A  a i1 Ai1  a i 2 Ai 2    ain Ain (cofactor expansion by row i).

A  a1 j A1 j  a 2 j A2 j    a nj Anj (cofactor expansion by column j).

The cofactor expansion by the first row is given by

Similarly the cofactor expansion by the second row is given by

The development of the determinant by the first column may be written as

It is easily verified that x = -3, -3 and 5 satisfy the above equation.

1. Evaluate the determinants:

1.5 COFACTOR METHOD FOR A-1

If A  0, we show that A 1 exists by constructing it as follows:

First we need to define the adjoint matrix.

 A11 A21 An1 

A11  9, A12   3, A13   9,

From (4) it follows that A 1 exists if and only if A  0.

If A 1 exists, that is A is invertible, then A is called a non-singular matrix. If A 1 does