MATRIX THEORY

UNIT 2: MATRIX THEORY

INTRODUCTION
In this chapter, we shall study the concept of a matrix, one of the fundamental tools of
mathematics. You may have seen matrices in connection with the solution of equations, and
we explain exactly how that application of matrices works. However, matrices have
fascinating applications in the theory of graphs, in geometry, in multivariable calculus, and
even in probability. Matrices also provide an excellent data structure for use in computer
programs. We shall see how the answers to certain practical questions can be found by
applying standard arithmetic operations on matrices. We begin our study of matrices with a
study of their arithmetic and some of the algebraic properties this arithmetic has. Next, we
make a careful study of the role of matrices in solving systems of equations. We develop a
technique that can be used by a person or programmed onto a computer to determine all
solutions to any system of linear equations. Our study of equation-solving leads us to the
concept of an invertible matrix and the inverse of such a matrix. This idea has intellectual
appeal, since it extends ideas from arithmetic and ordinary algebra, as well as vast practical
significance. Many practical applications of invertibility require more knowledge than we can
provide here. In order to determine whether matrices are invertible, we introduce the concept
of a determinant. The reader may have been introduced to the time-consuming and detailed
application of determinants to the solution of systems of two or three linear equations in two
or three unknowns. Since we have more efficient methods for solving equations, we do not
pursue this application, but rather the application of determining invertibility.
Objective: After successful completion of this unit you should be able to:

Identify different types of matrices;

Add , subtract and multiply matrices;

Multiply matrices by scalars;

Find the transpose a matrix;

Calculate the determinant of a matrix;

Describe properties of determinant;

BIT MATHEMATICS PROGRAM

40
 LINEAR ALGEBRA

Calculate the adjoint of a matrix;

Use properties to calculate determinant of a matrix;

Change matrices in reduced row echelon forms using elementary row operations;

Find the rank of matrices;

Find inverse of a matrix using elementary row operations ;

Write a given system of linear equations in matrix form;

Solve any system of linear equations using different methods;

2.1 MATRIX
Definition 2.1:
A rectangular array or arrangement of numbers of the form
 a11 a12 a13 . . . a1n 
a . . . a 2 n 
is called an
×  matrix, with
rows and  columns
 21 a 22 a 23
A =  a31 a32 a33 . . . a 3n 
 
 M M M M M 
a m1 am 2 a m3 . . . a mn 

uppercase letters: , , , and so on.

and the numbers are called the entries of the matrix. Matrices are usually denoted by



Hence,
 ,
 , ⋯ ,
  and  ⋮  represent respectively the  − ℎ row and the  − ℎ

column of the matrix , and
 represents the entry in the matrix  on the  − ℎ row and
 − ℎ column. We often write the matrix  as  = 
 × , ∀ = 1:
, ∀ = 1:  -denotes

the row and  −denotes the column.

1  has 3 rows and 4 columns. Therefore,

1 0 4 7
Example 2.1: the matrix A = 3 − 2 3
5 9 − 6 − 8

matrix A can be called a matrix of order or size 3 × 4.

BIT MATHEMATICS PROGRAM

41
 MATRIX THEORY

1
 
The rows are (1 0 4 7 ) , (3 − 2 3 1) and (5 9 − 6 − 8) , and the columns are  3  ,
 5
 
 0   4   7 
     
 − 2  ,  3  and  1  . Here a11 = 1 , a12 = 0 , a13 = 4 , a14 = 7 , a 21 = 3, a 22 = −2 , a 23 = 3 ,
 9   − 6  − 8
= 1 , etc. and  has 12 elements.
     
a 24

2.2 TYPES OF MATRICES

1. Row matrix: A matrix of size 1 ×  (or having exactly one row) is called a row matrix.
Example 2.2: (1 0 4 7 ) , (1 5) , are row matrices. A row matrix is often referred to as a

2. column matrix: A matrix having exactly one column(of size
× 1  is called a column
row vector

matrix
 4 
 
Example 2.3:  3  is a column matrix.
 − 6
 

and is denoted as 0.
3. A zero matrix or null matrix: A matrix in which every entry is zero is called the zero matrix

 0 0
 0 0  
Example 2.4:   ,  0 0  are the zero matrices
 0 0  0 0
 
4. A matrix, in which the numbers of rows and the number of columns are equal, that is, an  % 
matrix, is called a square matrix of order n.
 1 3 5
 4 0  
Example 2.5:   and  4 6 0  are square matrices.
 3 1  1 − 2 8
 
For a square matrix A = (aij )n×n the elements a ii ' s are called the diagonal elements. The

main diagonal, or simply diagonal of  consists of the elements  ,  , ⋯ , 

−4 2 3
Hence, the main diagonal extends from the upper left corner of A to the lower right

corner. For  = & 0 5 7*, the diagonal elements are −4,5 and 8.
1 10 8

BIT MATHEMATICS PROGRAM

42
 LINEAR ALGEBRA
5. A square matrix in which all the non-diagonal elements are zero is called a diagonal matrix.
 4 0 0
3 0  
Example 2.6:   and  0 − 5 0  are diagonal matrices.
0 2 0 0 9
 
Note that the diagonal elements in a diagonal matrix may also be zero.
0 0  0 0
Example 2.7:   and   are also diagonal matrix.
0 2  0 0
6. A diagonal matrix whose diagonal elements are equal is called a scalar matrix
 4 0 0
 3 0    0 0
Example 2.8:   ,  0 4 0  and   are scalar matrices
 0 3  0 0 4  0 0
 

matrix with 1’, on the main diagonal and zeros elsewhere) is called the identity matrix
7. A scalar matrix of order n in which every diagonal element is unity (one)(or is a square

or unit matrix of order n . The identity matrix of order n is denoted by I n .

1 0
Example 2.9: - = . / is an identity matrix of order two.
0 1
1 0 0
-0 = &0 1 0* is an identity matrix of order two.
0 0 1

8. A square matrix having only zeros below its diagonal is called upper triangular matrix. A
square matrix having only zeros above its diagonal is called lower triangular matrix.
A matrix that is either upper triangular or lower triangular is called triangular matrix.
1 0 0 1 2 0
  − 3 0   − 3 7
Example 2.10:  2 0 0  ,   are lower triangular matrices and  0 0 0  ,  
 0 4 3  6 4  0 0 3  0 4
   

2 0 0
are upper triangular matrix.

The matrix  = &0 1 0 * -is both upper and lower triangular matrix.
0 0 −5
Definition 2.2: Two matrices  and  are equal (written = ) if and only if :
1. They have the same size
2. Corresponding entries are equal.

BIT MATHEMATICS PROGRAM

43
 MATRIX THEORY

If the entries of  and  are written in the form  = 1

 2,  = 3
 , described earlier, then
the second condition takes the following form: 1
 2 = 3
  means
 = 3
 for all  and .
3 −4 1 6 7 3
Example 2.11: Given  = . /, = . / , 5  = . /, discuss the
0 5 3 0 −1 6 5
possibility that  = ,  = ,  = .
Solution:  =  is impossible because  and  are of different sizes:  is 2 × 2 whereas  is
2 × 3.Similarly,  =  is impossible. =  -is possible provided that corresponding entries
3 3 −4
are equal: . /=. / means = 3, 3 = −4, 6 = 0, and 5 = 5.
6 5 0 5
Definition 2.3: Let A be a matrix. A matrix obtained by deleting one or more rows or
columns of A is called sub matrix of A .
 4 6 − 1  4 6
  8 2   4 6  
Example 2.12: If A =  3 8 2  , then   ,   ,  3 8  are some of the sub
 2 0 − 3  0 − 3  3 8   2 0 
   
matrices of A

2.3 OPERATIONS ON MATRICES

2.3.1 ADDITION AND SUBTRACTION OF MATRICES

Definition 2.4:
If A and B are matrices of the same order, then sum of A and B , denoted by A + B , is the
new matrix of the same order obtained by adding the corresponding elements of A and B .
Similarly, the difference of A and B , denoted by A − B , is the matrix obtained by subtracting
the corresponding elements of A and B .
 1 3 5  4 6 − 1
   
Example 2.13 1: If A =  4 6 0  and B =  3 8 2  , then
1 − 2 8  2 0 − 3
   

 5 9 4 − 3 − 3 6 
   
i) A + B =  7 14 2  ii) A − B =  1 − 2 − 2 
 3 − 2 5  − 1 − 2 11 
   

BIT MATHEMATICS PROGRAM

44
 LINEAR ALGEBRA

2.3.2 MULTIPLICATION OF MATRICES BY SCALAR

Let A be any matrix and α be a scalar (real number), then αA is the matrix obtained from A
multiplying each element of A by α . This operation is called scalar multiplication.
In particular, − A is the matrix obtained from A by multiplying each element of A by -1and is
called the additive inverse of A .
3 −1 2  9 −3 6 
   
Example 2.14: If A =  5 0 4  , then 3 A = 15 0 8 
1 2 6  3 6 18 
   
Properties on Matrix addition and Scalar Multiplication

Let A, B and C be m × n matrices and O be a zero matrix of size m × n and α , β be scalars.

Then
1. A + B = B + A (Commutative law for addition)
2. ( A + B ) + C = A + (B + c ) (Associative law for addition)
3. A + O = O + A = A ( O is called Additive identity )
4. For each matrix A , there exists a unique m × n matrix - A such that
A + (− A) = O = − A + A

5. α ( A + B ) = αA + αB

6. (α + β )A = αA + βA
7. α (βA) = (αβ ) A = β (αA)
4 0 2 10 9 −1
Example 2.15: Given  = & 3 8* ,  = & 6 0 *, and  = & 0 1 *.
−2 5 −2 7 −2 0
Compute – ,  + , and  +  + 3.
−4 0 4 0 2 10 6 10
Solution: − = &−3 −8*,  +  = & 3 8* + & 6 0*=& 9 8*
2 −5 −2 5 −2 7 −4 12
9 −1 27 −3
3 = 3 & 0 1 * = & 0 3*
−2 0 −6 0
6 10 27 −3 33 7
 +  + 3 = & 9 8*+& 0 3 *=& 9 11*
−4 12 −6 0 −10 12

BIT MATHEMATICS PROGRAM

45
 MATRIX THEORY

2.3.3 MATRIX MULTIPLICATION

Matrix multiplication is a little more complicated than matrix addition or scalar multiplication,
but it is well worth the extra effort. It provides a new way to look at systems of linear

Definition 2.5: If A is an
×  matrix and B is an  × > matrix, the product AB of A and B

equations as we shall see, and has a wide variety of other applications.

is the
× > matrix whose ,  −entry is computed as follows:

Multiply each entry of row  of  by the corresponding entry of column  of , and add the
results. This is called the dot product of row  of  and column  of .
3
3
That is, if 
 ,
 , ⋯ ,
  is  − ℎ row of  and @  B is the  − ℎ column of the matrix
⋮
?3 A
, then the ,   −entry of the product matrix  is the dot product


 3 +
 3 + ⋯ +
 3 = C
D 3D
DE

Rule: Suppose  and  have sizes
×  and > × F respectively:

This requires that the rows of A and the columns of B be the same length.

 

×  > ×F

The product  can be performed only when  = >(the number of columns of the first matrix
must be equal to the number of rows of the second matrix); in this case, the product matrix 
is of size
× F. When this happens, we say that the product  is defined.

3 0 −1
1 −2 2
Example 2.16: Let  = . / ,  = &−2 1 4 *. Then compute .
4 3 0
0 5 6
Solution: The 1, 1 −entry of  is the dot product of row 1 of  and column 1 of ,

3
computed by multiplying corresponding entries and adding the results

1, −2,2 ⋅ &−2* = 1.3 − 2. −2 + 2.0 = 7

0
Similarly, the 1, 2 -entry of  involves row 1 of  and column 2 of .
0
1, −2,2 ⋅ &1* = 1.0 − 2.1 + 2.5 = 8.
5

BIT MATHEMATICS PROGRAM

46
 LINEAR ALGEBRA
Since  is 2 × 3 and  is 3 × 3, the product is 2 × 3.

3 0 −1
1 −2 2 7 8 3
. / &−2 1 4 *=. /
4 3 0 6 3 8
0 5 6
1 2
−2 −4 3 1
Example 2.17: Given  = &4 −1* ,  = . /. Then compute .
5 6 0 7
0 3
1 2 8 8 3 15
−2 −4 3 1
Solution: &4 −1* . / = &−13 −22 12 −3*
5 6 0 7
0 3 15 18 0 21

Note:
1. If  exists, does it happen that  exists and  = ? The answer is not necessary yes.
First  and  exist if and only if × , and × . Even if this is so the sizes of  and
 are different ( is
×
and  is  × ) unless
= . However even if
= 
we may have  ≠ . See the examples below. They may be different sizes and if they are
the same size (That is,. and  are square) the entries may be different.
2 2
Example 2.18:  = 3,1,  = . /.  = 3,1 . / = 10
4 4
2 6 2
 = . / 3,1 = . / ⟹  ≠ 
4 12 4
2 4 1 −3
=. / , = . /
1 5 −1 0

2 4 1 −3 −2 −6
 = . /. /=. /
1 5 −1 0 −4 −3

1 −3 2 4 −1 −11
 = . /. /=. /
−1 0 1 5 −2 −4

⟹  ≠ 

2. If  is square we define  =  ,  =  , 0 =   = 

 = J  =  ⋯   factors).
3. If - is identity matrix of order  and × , then - =  = -.( Identity matrices play a
neutral role with respect matrix multiplication)

BIT MATHEMATICS PROGRAM

47
 MATRIX THEORY

Properties of Matrix Multiplication

Let A, B and C be three matrices of the appropriate sizes. Let α be a scalar. Then
1. A(BC ) = ( AB ) C .
2. A(B + C ) = AB + AC and ( A + B )C = AC + BC .
3. (αA)B = α ( AB ) = A(αB )

2.4 THE TRANSPOSE OF A MATRIX

Definition 2.5: A matrix obtained from a given matrix A by interchanging the rows and
columns is called the transpose of A and it is denoted by A t .That is, If A = (aij )m×n then
A t = (a ji )n×m .
 2 5
   2 − 2 4
Example 2.19: 1. If A =  − 2 3  , then A t =  
 4 1 5 3 1
 
 4 
 
2. If B =  3  , then B t = (4 3 − 6 )
 − 6
2 1 3 2 1 3
 

3. If  = &1 0 −4*, then  = &1K

0 −4*
3 −4 −9 3 −4 −9

Definition 2.6 : Let  = 1

 2
×
be a square matrix. Then A is said to be

i) Symmetric matrix if A t = A
ii) Skew symmetric if A t = − A

 2 3 − 5  2 3 − 5
   
Example 1.20: 1. A =  3 0 2  , A =  3 0 2  , ⇒ A t = A . Therefore, A is
t

− 5 2 1  − 5 2 1 
   
symmetric.

 0 − 2  0 2
2. A =   , A t =   = − A .Therefore, A is skew-symmetric
2 0  − 2 0

BIT MATHEMATICS PROGRAM

48
 LINEAR ALGEBRA
Properties of Transpose of Matrix

Let A and B be matrices such that addition and multiplication is defined. Then
( )
1. At
t
=A
2. ( A + B ) = A t + B t
t

3. ( AB ) = B t A t
t

4. (αA) = αA t , α - is a scalar
t

1
(
1
5. A = A − A t + A + A t
2 424 2 424
) ( )
1 3 1 3
Skew Symmetric

Example 2.21: If  and  are symmetric  ×  matrices, show that  +  is symmetric.

Solution: We have K =  and  K = ,so, by property 2,  + K = K +  K =  + .
Hence  +  is symmetric.
Example 2.22: Show that  =  − K is skew-symmetric.
Solution:  K =  − K K = K −  = − − K  = −
Therefore,  − K is skew-symmetric.
10 1 3
Example 2.23: Let = & 5 −2 0 * . Then find a symmetric matrix  and a skew-
1 6 −4
symmetric matrix  such that  =  + .

20 6 4 10 3 2
Solution:

The symmetric matrix  is  =   +  K

= & 6 −4 6 * = & 3 −2 3 *
 

4 6 −8 2 3 −4

0 −4 2 0 −4 2
The skew-symmetric matrix  is  =   − K  =  & 4 0 −6* = & 4 0 −6*
 

−2 6 0 −2 6 0
1 0 2 5
Example 2.24: Find the matrix  if  + 2 . /=. /.
3 −1 3 −4
3
Solution: must be a 2 × 2 matrix. If = . / , the equation becomes
6 5
3 1 0 2 5
. / + 2. /=. /
6 5 3 −1 3 −4
3 2 0 2 5
. /+. /=. /
6 5 6 −2 3 −4
+2 3 2 5
⟹. /=. /
6+6 5−2 3 −4

BIT MATHEMATICS PROGRAM

49
 MATRIX THEORY

The rule of matrix equality gives + 2 = 2 ⟹ = 0

3=5
6 + 6 = 3 ⟹ 6 = −3
5 − 2 = −4 ⟹ 5 = −2
0 5
Thus,  = . /
−3 −2
1 2 5 0
Example 2.25: Find  if  . /=. /
3 4 7 2
3
Solution:  must be a 2 × 2 matrix. If = . / , the equation becomes
6 5
3 1 2 5 0 + 33 2 + 43 5 0
. /. /=. /⟹. /=. /
6 5 3 4 7 2 6 + 35 26 + 45 7 2
The rule of matrix equality gives + 33 = 5 ----------- (1)
2 + 43 = 0 --------- (2)
6 + 35 = 7 ----------- (3)
26 + 45 = 1--------- (4)
From (1) and (2), we get = −10, 3 = 5
From (3) and (4), we have 5 = 6 , 6 = −11
−10 5
Therefore,  = . /
−11 6

1 3
2 1 3 −1 2 3 −1
Exercise 2.1:

Let  = . / , = . / , = . / , L = &−1 0*, and

0 −1 0 1 4 2 0
1 4
1.

1 0 1
M=. /. Then compute the following (where possible)
0 1 0

a) 3 − 4 b)  + 2L c)  + K d) K

e) 5K − 6 f) L +  +  g) LK  + M h)  − LMK

3 6 − 35 −5
2. Find , 3, 6 ,and 5 if . /=. /
6 5 2 + 5 + 3

3. Find  if

1 0 5 2 1 2 7 2
a) 5 − . / = 3 − . / b) . / = . /
2 3 6 1 −1 2 −1 4

BIT MATHEMATICS PROGRAM

50
 LINEAR ALGEBRA
K 2 1
1 −1 0
6 N + 3 . /O = &0 5*
1 2 4
3 8
, 2, ,
4. Determine all , and  such that the given matrix is symmetric &  −1 , *
 , ,

2.5 DETERMINANT OF MATRIX

Definition 2.7: For each square matrix  of order n , we associate a number called

determinant of A .The determinant of A is denoted by det( A ) or A

If  =   − is a 1 × 1matrix. Then the determinant of A is defined as the scalar .

Determinant of a matrix of order one

Example 2.26: The determinant of  = 6 is 6 and the determinant of  = −5 is −5.
Determinant of order two

Let A =  11 12  be a square matrix of order two. Then we define the determinant of A as
a a
 a 21 a 22 

a 11 a 12
det( A ) = A = = a11 a 22 − a 21 a12 .
a 21 a 22

3 4
Example 2.27: If A =   , then A = 3 × 5 − 4 × 1 = 11
1 5
10 9
If = . / , then || = 107 − 09 = 70
0 7
Determinant of order three
 a11 a12 a13 
Let A =  a 21 a 22 23 

.Then
a a 32 
a 33 
 31

a11 a12 a13 a 21 a 22

a 22 a 23 a 21 a 23
det ( A ) = a21 a22 a23 = a11 − a12 + a13
a 32 a 33 a 31 a 33 a 31 a 32
a31 a32 a33

= a11 (a 22 a33 − a 32 a 23 ) − a12 (a 21 a33 − a 23 a 31 ) + a13 (a 21 a 32 − a 22 a 31 ) is the determinant of

BIT MATHEMATICS PROGRAM

51
 MATRIX THEORY

A expanded along the 1st row. We can find the determinant of A by using expansion in terms
of any of the three rows (columns) at which + or - sign can be determined using the pattern
+ − + 
− + −  .
 
+ − + 

1 2 3 1 2 2
 
Example 2.28: If A = − 2 4 5 ,then A = − 2 4 5
 7 − 1 6 7 −1 6

= 1 × (4 × 6 − (− 1) × 5) − 2 × (− 2 × 6 − 7 × 5) + 3 × (− 2 × −1 − 7 × 4) = 45
3 5 0
Example 2.29: Find the || for  = &−1 2 1*
3 −6 4

|| = 3 Q 2 1Q − −1 Q 5 0 5 0
Solution: We arbitrarily expand by the first column. Thus,

Q + 3Q Q
−6 4 −6 4 2 1
= 38 + 6 + 20 − 0 + 35 − 2
= 42 + 20 + 15 = 77

|| = 3 Q 2 1Q − 5 Q−1 1Q + 0 Q−1 2 Q

Or expanding by the first row

−6 4 3 4 3 −6
= 38 + 6 − 5−4 − 3 + 0 = 42 + 35 = 77
This example illustrates two important properties of expansion by cofactors.
First, the value of a determinant is the same regardless of which row or column selected and
second, expanding by a row or column containing zeros significantly reduces the number of
computations.
Determinant of n × n Matrix
 a11 a12 a13 . . . a1n 
a a 2 n 
 21 a 22 a 23 . . .
Let A =  a31 a32 a33 . . . a3n  be a matrix of size n × n . Let i be an integer such that
 
 M M M M M 
a m1 a m 2 a m3 . . . a mn 

1 ≤ i ≤ n . We define the determinant of A by

det( A) = (− 1) ai1 det( Ai1 ) + (− 1) ai 2 det( Ai 2 ) + (− 1) ai 3 det( Ai 3 ) + ... + (− ) ain det( Ain )
i +1 i+2 i +3 i +n

BIT MATHEMATICS PROGRAM

52
 LINEAR ALGEBRA
n
= ∑ (−1) i + j aij det( Aij ) , where Aij is a (n − 1) × (n − 1) matrix obtained by deleting

the ℎ row and the ℎ column of A.

j =1

Example 2.30: Find the determinant of the matrix

1 4 3 2
1 − 1 0 2 
A= .
5 1 − 2 − 3
 
0 1 2 4

Solution: det ( A) = a11 det ( A11 ) − a12 det ( A12 ) + a13 det ( A13 ) − a14 det ( A14 ) (expanded along first
row)
−1 0 2 1 0 2 1 −1 2 1 −1 0
=1 1 − 2 − 3 -4 1 − 2 − 3 +3 1 1 − 3 -2 1 1 −2
1 2 4 0 2 4 0 1 4 0 1 2

= 1 × 10 − 4 × 2 + 3 × 13 − 2 × 6 = 29
2 0 6 −2
Example 2.31: Find det  for  = T−2 5 1 0U
4 0 2 1
2 1 0 −3
Solution: The row or column containing the most zeros is, for this matrix, the second column,

−2 1 0 2 6 −2 2 6 −2 2 6 −2
so we expand by it.

|| = 0 V 4 2 1 V + 5 V4 2 1 V + 0 V−2 1 0 V + V−2 1 0V

2 0 −3 2 0 −3 2 0 −3 4 2 1
2 6 −2 2 6 −2
= 5 V4 2 1 V + V−2 1 0 V, and
2 0 −3 4 2 1
2 6 −2
2 1 4 1 4 2
V4 2 1 V = 2 Q Q − 6Q Q − 2Q Q (Expanding by first row)
0 −3 0 −3 2 0
2 0 −3
= 2−6 − 0 − 6−12 − 0 − 20 − 4 = −12 + 72 + 8 = 68
2 6 −2
1 0 −2 0 −2 1
V−2 1 0 V = 2Q Q − 6Q Q − 2Q Q(Expanding by first row)
2 1 4 1 4 2
4 2 1
= 21 − 0 − 6−2 − 0 − 2−4 − 4 = 2 + 12 + 16 = 30
Thus, det = 68 + 30 = 98

BIT MATHEMATICS PROGRAM

53
 MATRIX THEORY

Exercise 2.2: Compute the determinant of the following matrices

 1 2 3
 2 − 5  
a) A=  b) B =  2 0 0
9 4  −1 3 4 

5 0 11 − 5 0 
− 3 2 13 7 − 4 5 4 − 1 − 3
  
c)  0 1 5 0 0  d)  1 0 2 1 
  − 1 1 6 − 2
2 1 9 0 0  
0 0 − 5

 1 − 1 − 4 0 − 2 2

Note: With no zero entries, the determinant of an  ×  matrix requires ! multiplications.

Note that 10! = 3, 628, 000 and 14! is over 2 billion, so the number of multiplications
needed to evaluate a determinant becomes too expensive as the order of a matrix increases.
Clearly, calculating a determinant is a difficult and time-consuming process, one that is
avoided whenever possible. That is why the following properties of determinant are so
surprising.

2.5.1 PROPERTIES OF DETERMINANTS

1. If a matrix B is obtained by interchanging two rows (columns) of A , then
det(B) = − det(A) .
a1 a2 a3 b1 b2 b3 a2 a1 a3
That is, b1 b2 b3 = − a1 a2 a3 = − b2 b1 b3 .
c1 c2 c3 c1 c2 c3 c2 c1 c3
4 1 5 6
Example 2.32: Verify this property for  = . / and  = . /.
5 6 4 1
|| = Q4 1Q = 24 − 5 = 19, and || = Q5 6Q = 5 − 24 = −19
5 6 4 1
⟹ || = −||

2. If matrix B is obtained from matrix A by multiplying the elements of one of the rows or
columns by a scalar k , then det( B ) = k det( A) .

BIT MATHEMATICS PROGRAM

54
 LINEAR ALGEBRA

a1 a2 a1 a2
Example 2.33: If = 2 , then find B =
αb1 αb2
   
b1 b2

Solution: = QX3 X3 Q = X Q3 3 Q = 2X


If k is a real number and A is n × n matrix, then det( kA) = k det( A) .

n
3.

a1 a2 a3  αa1 αa 2 αa 3 
 
Example 2.34: If b1 b2 b3 = 5 , then find the determinant of  αb1 αb2 αb3 
c1 c2 c3  α c αc αc3 
X  X  X 0   0
 3 2

Solution: VX3 X3 X30 V = XXX V3 3 30 V = 5X 0

X6 X6 X60 6 6 60
4. If one of the rows (columns) of A is a scalar multiple of another, then det( A) = 0 .That is,

a1 a2
=0.
ka1 ka2
 1 2 − 3
 
Example 2.35: Find the determinant of A =  2 3 1 
 2 4 − 6
1 2 −3
 

Solution: || = V2 3 1 V = 0, because the third row is the scalar multiple of the first
2 4 −12
row.
5. If a matrix B is obtained from A by adding a multiple of a row (a column) to another row
 a1 a2 a3 
 
(Column), then det( A) = det(B ) .That is, if A =  b1 b2 b3  , then
c c2 c3 
 3
a1 + kb1 a 2 + kb2 a3 + kb3
b1 b2 b3 = A
c1 c2 c3

6. If a row (column) of a matrix is zero, then its determent is zero (expansion by a zero row or
zero column, when it exists, makes the process trivial.)
1 3 6
Example 2.36: 0 0 0 =0
−5 8 7

BIT MATHEMATICS PROGRAM

55
 MATRIX THEORY

7. The determinant of a matrix and its transpose are equal. That is, det ( A) = det A t ( )
a1 a2 a3 a1 b1 c1
That is, b1 b2 b3 = a 2 b2 c2 .
c1 c2 c3 a3 b3 c3

8. The determinant of a triangular matrix is the product of its diagonal elements.

a1 0 0 0
a2 a3 0 0
Example 2.37: = a1a3b3 c 4
b1 b2 b3 0
c1 c2 c3 c4

9. The determinant of a product of two matrices is the product of their determinants. That is,
( )
det ( AB ) = det ( A) det (B ). ⇒ det A n = (det ( A))
n

10. If two rows (columns) of A are equal, then det( A) = 0 .

1 5 −7
Example 2.38: 2 3 1 = 0 , because the first and last row of the matrix are equal.
1 5 −7

2 0 0
11. The determinant of a diagonal matrix is the product of its diagonal elements.

Example 2.39: V0 3 0 V
0 0 −12

2 0 0
Solution: Expanding by the first row,

3 0 0 0 0 1
V0 3 0 V = 2Q Q − 0Q Q + 0Q Q
0 −12 0 −12 0 0
0 0 −12
= 23−12 + 0 + 0 = 2 × 3 × −12 = −72. This is the product of the
diagonal elements.
Exercise 2.3:
1. Find the determinant of the following matrices:
4 0 0 0 0
10 2 0 12
@8 12 0 0 0B
a)  = T 6 3 0 1U b) = Y11 0 6 0 0Z
7 9 0 1
7 1 5 −1 0
0 −2 0 4
?3 2 5 −1 9A

BIT MATHEMATICS PROGRAM

56
 LINEAR ALGEBRA

2 0 6 −2
0 0 3 0 0
3 0 0 0
−2 5 1 0U
− 6 0 0 0 0
d)  = T
 0 2 0 0 
4 0 12 −4
c) C =  0 0 0 − 5 0 e) A = 
2 1 0 −3
  0 0 11 0 
0 0 0 0 9  
 0 0 0 0 4
1 0 0 0
> 2>
2. a) Find > so that Q Q = −12
4 >
b) If  and  are 3 × 3 matrices with || = 6, || = 3, then find |32K |
6 3
3. Given that V5 [ \ V = 8. Then find
]  ℎ
5 [ \ 4 43 46 +] 3+ℎ 6+
a) V 6 3V b) V−25 −2\ −2[ V c)^ ] ℎ  ^
]  ℎ 10] 10ℎ 10 −5 −\ −[

2.6 INVERSE OF A MATRIX

Definition 2.8: An n × n square matrix A is said to be invertible (non-singular) if there
exists an n × n matrix B such that AB = I n = BA .The B is called inverse of A and it is

denoted by A −1

1 2 7 −2
If A does not have an inverse, A is called singular (non-invertible).

Example 2.40: Let = . / , and  = . /. Then

3 7 −3 1
1 2 7 −2 1 0 7 −2 1 2 1 0
 = . /. /=. /, and  = . /. /=. /
3 7 −3 1 0 1 −3 1 3 7 0 1
1 0 1 0
⟹  = . / and  = . /
0 1 0 1
Then we conclude that B is the inverse of A and J = .
 Inverse of invertible matrix is unique.

 If A is invertible, then det( A) ≠ 0 and det (A −1 ) =

1
det ( A)
1 2 − 2 1 
Example 2.41: A =   , then det( A) = −2 and A −1 =  3 −1 ,
3 4  
 2 2 
−1
det (A −1 ) =
1
= .
2 det ( A)

BIT MATHEMATICS PROGRAM

57
 MATRIX THEORY

Properties of the inverse

1 .If A is nonsingular matrix, then A −1 is nonsingular and A −1 ( )−1
=A

2. If A and B are nonsingular matrices, then AB is nonsingular and ( AB ) = B −1 A −1

−1

3. If A is nonsingular matrix, then At ( ) −1

( )
= A −1
t

4 1 −1
Example2. 42: Given  = &10 6 3 *.Then || = 28 ≠ 0.
0 0 2
Therefore,  is invertible.
How does one can find the inverse of an invertible matrix?
2.6.1 ADJOINT-DETERMINANT METHOD
MINORS, COFACTORS AND ADJOINT

Definition 2.10: Let A = (aij ) be a square matrix of order n and further let 
 be a square

sub-matrix of  obtained by deleting  − ℎ row and  − ℎ column of A . Then

1. det (Aij ) is called the minor of the element aij .

2. C ij = (− 1) det (Aij ) is called the cofactor of aij .

i+ j

3. The transpose of the matrix of the cofactors C = (C ij ) is called adjoint of A and

denoted by adj ( A) .

C11 C12 L C1n 

C 
If [C ]
ij n×n =
 M
21 C
M
22 L
M
C 2 n 
M  is the matrix of cofactors, then
 
C n1 C n 2 L C nn 

C11 C12 L C1n  C11 t C 21 L C n1 

C  C C 22 L C n 2 
[ ]
Adj ( A) = Cij t
n×n =  21
 M
C
M
22 L
M
C 2n   12
M  =  M M M M 
   
C n1 C n 2 L C nn  C1n C 2 n L C nn 
1 − 2 2 
Example 2.43: Let A = 4 1 − 3 . Find the adjoint of A .
2 1 1 

BIT MATHEMATICS PROGRAM

58
 LINEAR ALGEBRA
1 −3 4 −3 4 1
Solution: C11 = =4 C12 = − = −10 C13 = = 2;
1 1 2 1 2 1
−2 2 1 2 1 −2
C21 = − =4 C22 = =0 C23 = − = −5 ;
1 −3 2 1 2 1
−2 2 1 2 1 −2
C31 = =4 C32 = − = 11 C33 = =9
−31 4 −3 4 1
t
4 − 10 2   4 4 4

Adj (A) = 4 0 − 5 = − 10 0 11
 

4 11 9   2 − 5 9 
10 2
Example 2.44: Given  = . /.Then find the adjoint of .
5 1
Solution: 6 = 1 , 6 = −5, 6 = −2, and 6 = 10
1 −5
Cofactor matrix  of  is  = . /, and hence the adjoint of  is
−2 10
1 −5 K 1 −2
Adjoint =  K = . / =. /
−2 10 −5 10

Exercise 2.4: Find the adjoint of the following matrices:

9 1 5
 = &1 6 0 *
 2 3 −1
 
5 2 7
a) b) A =  0 − 4 2 
 1 −1 5 
3 6 −9 0 3 1
 

c)  = & 0 0 −2* d)  = &1 1 2*

−2 1 5 3 2 4

DETERMINANT AND INVERSE

Let A = (aij ) be n × n square matrix. Then A is invertible if and only if A ≠ 0 and for this
case
1
A −1 = adj ( A)
A
1 2 3 
Example 2.45: Find the inverse of A = 4 6 5  using adjoint method.
7 8 − 1

Solution: A = 1(6 × (−1) − 8 × 5) − 2(4 × (− 1) − 7 × 5) + 3(4 × 8 − 7 × 6) = 2 ≠ 0 .

BIT MATHEMATICS PROGRAM

59
 MATRIX THEORY

Hence, A is invertible.
6 5 4 5 4 6
C11 = (−1)1+1 = −46 , C12 = (−1)1+ 2 = 39 , C13 = (−1)1+3 = −10
8 −1 7 −1 7 8

2 3 1 3 1 2
C 21 = (−1) 2+1 = 26 , C 22 = (−1)1+1 = −22 , C 23 = (−1) 2+3 =6
8 −1 7 −1 7 8

2 3 1 3 1 2
C 31 = (−1) 3+1 = −46 , C 32 = (−1) 3+ 2 = 7, C 33 = (−1) 3+ 3 = −2
6 5 4 5 4 6
t
− 46 39 − 10 − 46 26 − 8 

adj ( A) =  26 − 22 6  =  39 − 22 7 

 − 8 7 − 2   − 10 6 − 2

− 46 26 − 8  − 23 13 − 4
A = adj ( A) =  39 − 22 7  =  392
1   − 11 72 
1
−1

A 2
 − 10 6 − 2  − 5 3 − 1 
− 23 13 − 4 1 2 3  1 0 0
Check: A −1 A =  392 − 11 72  4 6 5  = 0 1 0
 − 5 3 − 1  7 8 − 1 0 0 1

4 −3
Example 2.46: Find the inverse of the matrix

=. /
2 1
Solution: 6 = 1 , 6 = −2, 6 = 3, and 6 = 4
1 −2
Cofactor matrix  of  is  = . /, and hence the adjoint of  is
3 4
1
−2 K 1 3
Adjoint =  K = . / =. /, and || = 10 ≠ 0.
4 3 −2 4
1 3
Therefore, -is invertible and hence the inverse of  is J = . /.

_ −2 4

3
In general, the matrix  = . / is invertible if and only if 5 − 63 ≠ 0, in which case
6 5
5 −3
J = `aJbc . /

−6

BIT MATHEMATICS PROGRAM

60
 LINEAR ALGEBRA
Exercise 2.5: Find the inverse of the following matrices using adjoint method.
0 1 1  1 1 1 − 2 2 − 4
a) 2 3 4 b) 1 2 3 c)  3 0 1 
1 0 1 1 4 9  1 − 2 2 

2.7 ELEMENTARY ROW OPERATIONS

Definition 2.11:
Given any matrix A of order m × n .Any one of the following operations on the matrix is
called elementary row operation.

Ri ↔ R j (Interchange row  and )

1. Interchanging any two rows of A .

2. Multiplying a row of A by a nonzero constant k

Ri → kRi (Multiply row  by scalar k )
3. Adding a multiple of one row of A to another row of A .
R j → R j + kRi (add > times  Kd row to Kd row).

 2 1 0 
Example 2.47: Perform elementary row operations on the matrix A = − 4 2 7  .
 1 9 0 

 2 1 0  1 9 0
R ←→ R
Solution: − 4 2 7 1 3
   →
− 4 2 7 
 
 1 9 0  2 1 0

 1 9 0 1 9 0  1 9 0
− 4 2 7 R2 → R2 + (4 R1 ) 0 38 7 R3 → R3 +(−2) R1 0 − 2 7 
        →
       →  
 2 1 0 2 1 0 0 − 18 0

1 9 0 1 9 0 
0 − 2 7 R3 → R3 + (−9 R2 ) 0 − 2 7 
        →

0 − 18 0 0 0 − 63

BIT MATHEMATICS PROGRAM

61
 MATRIX THEORY

Definition 2.12: (Equivalent matrices)

Two matrices A and B are called row equivalent if one can be obtained from the other by a
(finite) sequence of elementary operations. In this case we write A ~ B .
 2 1 0  1 9 0
Example 2.48: As we observe from the above example, − 4 2 7 ~ − 4 2 7 ,
 1 9 0  2 1 0
 2 1 0  1 9 0   2 1 0 1 9 0 
− 4 2 7  ~ 0 38 7 , and − 4 2 7 ~ 0 − 2 7 
      
 1 9 0 2 1 0  1 9 0 0 0 − 63
Definition 2.13 :( Reduced row echelon form)
A matrix with
rows and  columns (called an
×  matrix) is in row-reduced form if it satisfies the
following conditions.
1. All zero rows of A (if any) are at the bottom of the matrix.
2. The first non zero entry in any non zero row is 1 . This entry is called leading one
3. The first non zero entry in any row appears in a column to the right of the first non zero
entry in any preceding row.

Remark: An m × n matrix  satisfying properties 1, 2 and 3 is said to be in row echelon

4. If a column contains a leading one, then all other entries in that column are zeros.

form.
Example 2.49: Determine which of the following matrices are in row-reduced form. If a
matrix is not in row-reduced form, state the condition that is violated.

1 0 6 0
1 0 0 3 0 1 5 0 0 0 1 1 0
a) 0 1 0 − 4 1 b) 0 0 0 1 c) 
0 0 0 1
0 0 1 2 1 0 0 0 0  
0 0 0 0

1 0 0 0
0 1 2 0 0 1 1 0 0 
1 0 0
d)  e) 0 1 0 9 f) 0 0 1 − 1
0 0 1 0
  0 0 12 0 0 0 0 1 
0 0 0 1

BIT MATHEMATICS PROGRAM

62
 LINEAR ALGEBRA

0 1 − 4  1 1 0 0
g) 1 0 0  h) 0 0 0 0 
 
0 0 1  0 0 0 1

e) This matrix is not in row-reduced form. Conditions 2 and 4 are violated. The first
Solution: The matrices in parts (a)-(d) are in reduced-row echelon (row-reduced) form

nonzero entry in row 3 is a 2, not a 1.Also, column 2 contains a leading 1 and has

f) This matrix is not in row-reduced form. Condition 4 is violated. Column 4 contains

a nonzero entry above it.

a leading 1 in row 3 and a nonzero element above it.

g) This matrix is not in row-reduced form. Conditions 3 and 4 are violated: The
leading 1 in row 2 lies to the left of the leading 1 in row 1. Also, column 3
contains a leading 1 in row 3 and a nonzero element above it.
h) This matrix is not in row-reduced form. Condition 1 is violated: Row 2 consists
of all zeros and does not lie below the nonzero rows.

Example 2.50: Determine which of the following matrices are in row-echelon form. If a
matrix is not in row-echelon form, state the condition that is violated
1 9 0 1 5 0 0 3 2 0 4
a) 0 1 7 b) 0 0 1 6 c) 0 0 − 4 − 7

0 0 0 0 0 0 1 0 0 0 1 

1 6 −3 1
0 1 4 − 8
− 8
e) 0 0 0 
0 1
d) 
0 1 12 20 
  0 1 10 
0 0 0 5

c) This matrix is not in row-echelon form. Condition 2 is violated. The first nonzero
Solution: The matrices in parts (a) and (b) are in row echelon form.

entries in row 2 and 3 are not leading 1.

d) This matrix is not in row-echelon form. Conditions 2 and 3 are violated: The
leading 1 in row 3 lies to the left of the leading 1 in row 2. The first nonzero entry
in row 4 is 5, not 1.

BIT MATHEMATICS PROGRAM

63
 MATRIX THEORY

e) This matrix is not in row-reduced form. Condition 1 is violated: Row 2 consists

of all zeros and does not lie below the nonzero rows.
Exercise 2.6: Indicate whether the matrix is in row-reduced form or in row-echelon form.
1 0 6 0
0 1 1 0
1 0 4   0 1 3 0 1 3 0 7
a)   c)   e)   h) 
 0 1 − 5 1 0 7  0 0 0 1 0 0 0 1 1
 
0 0 0 0

1 0 4 0 7
  0 0 0 0  1 0 0 0 
 1 1 − 5 0 1 1 0 6
b)   d)  f ) 0 1 − 4 − 7  g ) 0 1 0 − 7 
  
0 0 0  0 0 0 1 9
  0 0 0 1  0 0 − 1 1 
0 0 0 0 0 

Remarks:
1. Any m × n matrix A can be changed to row echelon form or reduced row echelon form
by applying appropriate sequence of elementary operations.
2. Different sequences of elementary row operations starting with same matrix A may yield
different row echelon forms of A .
3. Any different sequences of elementary row operations starting with same matrix A
always yield one and only one reduced row echelon form of A . This means, the reduced
echelon form of a matrix A is unique.
Example 2.51: Find the reduced row echelon form of the matrix
1 3 −2 0 0 2 0
 
2 6 − 5 − 2 4 − 3 − 1
A=
0 0 5 10 0 15 5 
 
2 8 4 18 6 
 6 0
Solution: Adding −2 times the first row to the second and fourth rows gives

1 3 −2 0 20 0
 
0 0 − 1 − 2 0 − 3 − 1
0 0 5 10 0 15 5 
 
0 8 0 18 6 
 0 4
Multiplying the second row by -1 and then adding -5 times the new second row to the third
row and -4 times the new second row to the fourth row gives

BIT MATHEMATICS PROGRAM

64
 LINEAR ALGEBRA

1 3 − 2 0 2 0 0
 
0 0 1 2 0 3 1
0 0 0 0 0 0 0
 
0 0 0 0 0 6 2 


Interchanging the third and fourth rows and then multiplying the third row of the resulting

e
matrix by gives the row echelon form

1 3 − 2 0 2 0 0
 
0 0 1 2 0 3 1
0 0 0 0 0 1
1
 3
0 0 0 0 0 0 0 


Example 2.52: Using elementary row operations, transform the following matrix in to
reduced row echelon form:
2 0 − 6 − 8 1 0 − 3 − 4 1 0 − 3 − 4
  R → 1R   R3 → R3 −3R1  
a) A = 0 1 2 3  2
 2→
1
0 1 2 3    →  0 1 2 3 
 3 6 − 2 − 4  3 6 − 2 − 4 0 6 7 8 
    

1 0 − 3 − 4  1 0 − 3 − 4  R → − 1R
  R3 → R3 −6 R2   3 3

0 1 2 3     → 0 1 2 3   5 →
0 6 7 
8   0 0 − 5 − 10 
  

R2 → R2 − 2 R3
1 0 − 3 − 4 1 0 0 2 
  R1→ R1 + 3 R3  
0 1 2 3    →  0 1 0 − 1
0 0 1 2  0 0 1 2 
  

Therefore, the reduced row echelon form A is

1 0 0 2 
 
A =  0 1 0 − 1
0 0 1 2 
 

BIT MATHEMATICS PROGRAM

65
 MATRIX THEORY

1 1 − 1 3
 1 − 1 3 1
Example 2.53: Transform A =   to row echelon form by applying suitable
 3 5 − 7 1
 
2 1 2 3
row operation.

1 1 − 1 3 1 1 − 1 3 
1 − 1 3 1  R2 → R2 − R1 0 − 2 4 − 2 
Solution:   R → R3 −3 R1
  −1
 3 5 − 7 1 3 0 2 − 4 − 8  R2 → R2
     2 →
2 1 2 3 R4 → R4 − 2 R
 0 − 1 4 − 3 
  1 →
−1
R3 → R
R3 → R − 2 R 10 3
1 1 − 1 3  3 2 1 1 −1 3 
0 1 − 2 1  R4 → R4 + R 0 1 − 2 1 
1
R4 → R4
     2→    2→
0 2 − 4 − 8  0 0 0 − 10
   
0 − 1 4 − 3 0 0 2 −2

1 1 −13 1 1 −1
3
0 1 − 2 1  R3 ↔ R4 0 1 − 2 1 
   
→ 
0 0 0 1 0 0 1 − 1
   
0 0 1 − 1 0 0 0 1

1 1 −1
3
0 1 − 2 1 
Therefore, the row echelon form of A is 
0 0 1 − 1
 
0 0 0 1

 3 − 2 8 9
 
Example 2.54: Find the reduced row echelon form of the matrix A =  − 2 2 1 3
 1 2 − 3 8 


R3 → R + 2 R
3 2
 3 − 2 8 9  1 0 9 12 
  R1 → R1 + R2   R3 → R3 − R
Solution:  − 2 2 1 3     →  − 2 2 1 3    1→
 1 2 − 3 8   1 2 −3 8 
  

BIT MATHEMATICS PROGRAM

66
 LINEAR ALGEBRA

1 0 9 12  1 0 9 12 
  R ↔ R2  
 0 2 19 27  3  →  0 2 − 12 − 4  R → 1 R
 0 2 − 12 − 4   0 2 19 27   2 2
2 →
   

 1 0 9 12   1 0 9 12 
  R3 → R3 − 2 R2  
 0 1 − 6 − 2    →  0 1 − 6 − 2  R → 1 R
 0 2 19 27   0 0 31 31  
3


3
31→
   

R1 → R −9 R
1 3
 1 0 9 12   1 0 0 3
  R2 → R2 + 6 R3  
 0 1 − 6 − 2    →  0 1 0 4 
0 0 1 1  0 0 1 1
  

Exercise 2.7: Transform the following matrix in to reduced row echelon form
12 −4 1   1 1 −1 3
 1 
a) A =  1 0 − 7  b) A =  − 1 −1 2
 − 3
 6 −4 −7 1
0 9   
− 2 0 1 6

1 2 3  0 0 − 2 0 7 12 
 
c) A =  4 5 6  d ) A =  2 4 − 10 6 12 28 

 7 9   2 4 − 5 6 − 5 − 1
8  

2.7.1 RANK OF A MATRIX

Definition 2.14:
The rank of a matrix A , denoted by rank ( A ), is the number of nonzero rows remaining after
it has been changed into row echelon or reduced row echelon form.
Remark: If A is zero matrix then rank ( A ) is 0 .
Example 2.55: Determine the rank of the following matrices.
1 0 0 1 
3 1 2 3 − 1 
1 2 6 
a) A =  b) B = 3 6 9 − 3
− 1 2 5 − 4
  2 4 6 − 2
2 3 7 2
Solution:

BIT MATHEMATICS PROGRAM

67
 MATRIX THEORY

1 0 0 1  1 0 0 1 1 0 0 1  1 0 0 1
3 1 2 6  0 1 2 3  0 1 2 3  0 1 2 3 
a)  ~ ~  ~ ~
− 1 2 5 − 4 0 2 5 − 3 0 0 1 − 9  0 0 1 − 9
       
2 3 7 2  0 3 7 0 0 0 1 − 9  0 0 0 0

1 0 0 1
0 1 0 21 

0 0 1 − 9
 
0 0 0 0
Hence, rank( A ) = 3
1 2 3 − 1 1 2 3 − 1
b).  3 6 9 − 3  ~ 0 0 0 0 
   
2 4 6 − 2 0 0 0 0 

Therefore, rank ( B ) = 1

Exercise 2.8: Find the rank of the following matrices:

 − 12 −4 1   1 1 −1 3
 1 
a) A =  1 0 − 7  b) A =  − 1 −1 2
 − 3
 − 6 −4 −7 1
0 9   
− 2 0 1 6

1 2 3  1 1 − 2 0 7 12 
 
c) A =  1 6 6  d ) A =  − 2 − 4 − 10 6 12 28 

 7 9   2 4 − 5 6 − 5 − 1
0 

2.7.2 INVERSE OF A MATRIX USING ELEMENTARY ROW

OPERATIONS
To find A −1 if it exists we can proceed as follows
Step 1: form a partitioned matrix [ A I ] augmented matrix

Step 2: Apply elementary row operations to reduce [ A I ] to [ I C ] if the reduction can be

done then A −1 = C , otherwise A −1 doesn’t exist.

BIT MATHEMATICS PROGRAM

68
 LINEAR ALGEBRA

3 6 
Example 2.56: Let A =   . Find A −1 using elementary row operations.
1 4
Solution:
3 6 1 0 R ↔R
1  2 → 1 4 0 1 R →R −3R 1 4 0 1 
       221→
 
1 4 0 1 3 6 1 0 0 − 6 1 − 3
 2 
R → − 1R 1 4 0 1 R →R −3R  1 0 − 1 
2 6
   
2
→  1 11 2→  3
1

0 1 − 6
1

 2  0 1 − 16 
 2 
2 
 − 1
Therefore, A −1 =  3
−1 1 
 
6 2

1 0 1
Check that AA −1 = A −1 A = I 2

Example 2.57: Find the inverse of the matrix &2 1 1*

2 1 2
R3 → R − 2 R
3 1
1 0 1 1 0 0 1 0 1 1 0 0
  R2 → R2 − 2 R   R ↔R
Solution:  2 1 1 0 1 0     1 →
 0 1 − 1 − 2 1 0  2 
3→
2 1 2 0 0 1 0 1 0 − 2 0 1
   

1 0 1 1 0 0 1 0 1 1 0
0
  R →R − R   R → − 1R
0 1 0 − 2 0 1 33 2→ 0
 1 0 − 2 0 1  3  3→
0 1 −1 − 2 1 0  0 0 − 1 0 1 − 1
   

1 0 1 1 0 0 1 0 0 1 1 − 1
  R →R −R  
0 1 0 − 2 0 1  11 3→  0 1 0 − 2 0 1
0 0 1 0 − 1 1  0 0 1 0 −1 1 
1 1 −1
  

Therefore, J = &−2 0 1*
0 −1 1

BIT MATHEMATICS PROGRAM

69
 MATRIX THEORY

 2 0 − 4 − 10 
 
 0 3 3 0 
Example 2.58: Find the inverse of the matrix A = 
0 0 1 0 
 
− 3 1 
 2 3
1
R1 → R
 2 1
1 
 2 0 − 4 − 10 1 0 0 0   1 0 −2 −5 0 0 0
  1  2 
 0 3 3 0 0 1 0 0  R2 → 3 R2 0 1 1 0 1
Solution:    →  0 0 0
0 0 1 0 0 0 1 0  0 0 1 0 3 
   0 0 1 0
− 3 2 3 1 0 0 0 1  − 3 2 3 1 

 0 0 0 1

 1   1 
1 0 −2 −5 2
0 0 0 1 0 −2 −5 2
0 0 0
   
1 0 1
R →R +3R
441→
0 1 1 0 0  R →R − 2
0 0 44 R
2→
1 1 0 0 0 0
 3   3 
0 0 1 0
0 0 1 0 0 0 1 0
0 0 1 0
0 2 − 3 − 14 3  0 0 − 5 − 14 3 −2 
 0 0 1  0 1
 2   2 3 

 1   1 
1 0 −2 −5 2
0 0 0 1 0 −2 −5 2
0 0
0
   
0 1 1
1 1 0 0 0 0 R →R +5R
442→
0 1 1 0 0 0 0
 3   3 
0 0 1 0
0 0 1 0 0 0 1 0
0 0 1 0
0 0 − 5 − 14 3 −2  0 0 0 − 14 3 −2 
 0 1  5 1
 2 3   2 3 

 1  R2 → R − R  1 
1 0 −2 −5 2
0 0 0 2 3 1 0 −2 −5 2
0 0 
0
  −1  
0 1  R4 → R4 0 1
1 1 0 0 0 0   14→ 
1 0 0 0 −1 0 
 3  3 
0 0 1 0
0 0 1 0 0 0 1 0
0 0 1 0 
0 0 0 − 14 3 −2  0 0 0 1 −3 1 −5 1
 5 1  − 
 2 3   28 21 14 14 

 1   1 
1 0 −2 −5 2
0 0 0  1 0 0 −5 2
0 2 0 
   
0 1 1
1 0 0 0 −1 0  R →R +2R
113→
0 1 0 0 0 −1 0 
 3   3 
0 0 1 0
0 0 1 0  0 0 1 0
0 0 1 0 
0 0 0 1 −3 1 −5 1 0 0 0 1 −3 1 −5 1
 −   − 
 28 21 14 14   28 21 14 14 

BIT MATHEMATICS PROGRAM

70
 LINEAR ALGEBRA

 1   −1 5 3 −5 
1 0 0 −5 2
0 2 0  1 0 0 0 28 
   21 14 14 
0 1 1
1 0 0 0 −1 0  R →R +5R
11 4→

0 1 0 0 0 −1 0 
 3   3 
0 0 1 0
0 0 1 0  0 0 1 0
0 0 1 0 
0 0 0 1 −3 1 −5 1 0 0 0 1 −3 1 −5 1
 −   − 
 28 21 14 14   28 21 14 14 

 −1 5 3 −5 
 
 28 21 14 14 
Therefore, J =  0
 1
−1 0 
3 
 0 0 1 0 
−3 1 −5 1
 − 
 28 21 14 14 

Exercise 2.9: Using elementary row operations, find the inverse of the following matrices:

 1 −3 2   1 1 −1 3
A = − 2 2 − 2
 1 
a) b) A =  − 1 −1 2
 2 − 4 4  − 3 −4 −7 1
 
− 2 0 1 6

1 0 0 0
 
1 3 3 0 1 2 
c) A =  d) A =  
0 2 1 0 8 − 1
 
2 2 3 1 


2.8 SYSTEMS OF LINEAR EQUATIONS

The method of substitution used in lower grades is well suitable to solving a system of
linear equations when the number of linear equations and variables is small. But for
large systems, the steps involved in the procedure become difficult to manage.
In this section we will develop a systematic procedure for solving systems of linear equations.
The procedure is based on the idea of performing certain elementary row operations on the
augmented matrix for the system that simplifies it to a form from which the solution of the
system can be easily calculated.

BIT MATHEMATICS PROGRAM

71
 MATRIX THEORY

A general linear system of m equations in the n unknowns % , % , ⋯ , % can be written as

a 11 x1 + a 12 x 2 + a 13 x 3 + ... a 1 n x n = b1
a 21 x1 + a 22 x 2 + a 23 x 3 + ... + a 2 n x n = b 2 (*)
.
a m1 x1 + a m 2 x 2 + a m 3 x 3 + ... + a mn x n = b n

A solution of a linear system in  unknowns % , % , ⋯ , % is a sequence of  numbers

, , , , ⋯ , , for which the substitution % = , , % = , , ⋯ , % = , makes each equation a
true statement. These solutions can be written more briefly as , , , , ⋯ , ,  in which the

geometrically as points in -dimensional space.

names of the variables are omitted. This notation allows us to interpret these solutions

Note: we say that a linear system is consistent if it has at least one solution and inconsistent if
it has no solutions. Thus, a consistent linear system of equations has either one solution or
infinitely many solutions—there are no other possibilities. That is, every system of linear
equations has no, one, or infinitely many solutions. There are no other possibilities.
 x1 
 a 11 a 12 a 13 ... a 1 n  x   b1 
   2 b 
 a 21 a 22 a 23 .... a 2 n  , X = .  , and  2  ,these equations become the single
If A =   b = . 
 
.  .
 
 
 a m 1 a m 2 a m 3 .. a mn  . 
 xn  b n 
   

matrix equation f = 3.This is called the matrix form of the system of equations.
 a 11 a 12 a 13 ... a 1 n 
 
The matrix a a 22 a 23 .... a 2 n  is called the coefficient matrix of the system (*)
A =  21 
. 
 a m 1 am2 a m 3 .. a mn 

 x1 
 
The column matrix X =  x 2  is the matrix of unknowns.
 
 
 xn 
 b1 
b 
The column matrix  2 is the matrix of constants.
b = . 
 
. 
b n 
 

BIT MATHEMATICS PROGRAM

72
  ⋯  3
LINEAR ALGEBRA

The matrix |3 = & ⋮ ⋱ ⋮ V ⋮ * is called the augmented matrix of the linear
 ⋯  3
system (*) since it is obtained by joining the matrix of coefficients to the column matrix of
constants. The vertical line separates the column of constants from the matrix of coefficients.
Remark: A system of linear equations is said to be homogeneous if the constant terms are all
zero, otherwise, non homogenous.
Theorem 2.1: If [ A b] and [C d ] are row equivalent, then the systems AX = b and CX = d

have exactly the same solutions.

systems have % = 0, % = 0, ⋯ , % = 0 as a solution. This solution is called the trivial

Remark: Every homogeneous system of linear equations is consistent because all such

solution; if there are other solutions, they are called nontrivial solutions.
Theorem 2.2: If A is an m by n matrix then the equation AX = O has non trivial solution
only when rank ( A) < n otherwise, that is, rank ( A) = n then the trivial solution is unique.

Theorem 2.3: Consider an m equations with n variables AX = b then

a) If rank( [ A b] ) = rank( A ) = r = n then the system has unique solution.

b) If rank( [ A b] ) = rank( A ) = r < n then the system has infinitely many solutions.

c) If rank( [ A b] ) ≠ rank( A ) then the system has no solution.

2.8.1 SOLUTION OF SYSTEM OF LINEAR EQUATIONS

There are different methods that are developed for solving system of linear equations. Some
of these are:
1. Gaussian Elimination
2. Gauss-Jordan Elimination
3. Matrix inversion Method and
4. Cramer’s Rule.

BIT MATHEMATICS PROGRAM

73
 MATRIX THEORY

2.8.1.1 GAUSSIAN ELIMINATION METHOD

Procedure for solving the system of linear equation f = 3 using Gaussian Elimination as
follows:
Step 1: Form the augmented matrix [ A b]

Step 2: Transform [ A b] to row echelon form [C d ] using row operations.

Step 3: Solve the system CX = d by back substitution.

Example 2.59: Solve the system of linear equation:

% + 2% + %0 + %h = 1
2% + 5% + 2%0 + 2%h = 3
3% + 6% + 4%0 + 4%h = 5
−% − 2% − 2%0 + %h = 6
1 2 1 1 1
 
 2 5 2 2 3
Solution: The augmented matrix of the system is 
3 6 4 4 5
 
−1 − 2 − 2 1 6 


Applying, elementary row operations on this matrix to change into its echelon form

1 2 1 1 1 1 2 1 1 1
  R2 → R2 − 2 R1  
 2 5 2 2 3 0 1 0 0 1
 3 6 4 45  R → R3 −3 R1
3
0 0 1 1 2
   
−1 − 2 − 2 1 6  R4 → R4 + R1 0 0 − 1 2 7 
   → 

1 2 1 1 1 1 2 1 1 1 1 2 1 1 1
    1  
0 1 0 0 1 R →R + R 0 1 0 0 1  R4 → R4  0 1 0 0 1
0 443→ 3 →

0 1 1 2 0 0 1 1 2 0 0 1 1 2
     
0 0 − 1 2 7  0 0 0 3 9  0 0 0 1 3 
  

Science i > . [ A b] / = i >1 A 2 = 4 =  the solution exists and is unique.

Hence the above equation is reduced to

% + 2% + %0 + %h = 1
% =1
%0 + %h = 2
%h = 3

BIT MATHEMATICS PROGRAM

74
 LINEAR ALGEBRA
From this we get %h = 3 , and using back substitution we have %0 = −1 , % = 1, and
% = −3 Hence % , % , %0 , %h  = −3,1, −1,3 is the solution of the system.

Example 2.60: Solve the system

% + 5%0 = −4
% + 4% + 3%0 = −2
2% + 7% + %0 = −2

0 1 5 −4
Solution: The augmented matrix of the system is &1 4 3V−2*
2 7 1 −2

Using the Gaussian–elimination method, we obtain the following sequence of equivalent

0 1 5 −4 1 4 3 −2
augmented matrices:

&1 4 3V−2*  2 → &0 1 5V−4*

R ↔R

2 7 1 −2 2 7 1 −2
1 


1 4 3 −2 1 4 3 −2
&0 1 5 V−4* 33 2 → &0 1 5V−4*
R →R −2R R →R + R

0 −1 −5 2 0 0 0 −2
331→ 

The system is inconsistent (has no solution), because the last row would require that 0 = −2
if there were a solution. The solution set is empty.

% − 3%0 + 8%h = 1
Example 2.61: Use Gauss elimination method, to solve the linear system

2% + 2% + 9%0 + 7%h = 3
% + 5%0 − 2%h = 3

 1 0 − 3 8 1
 
Solution: The augmented matrix of the system is  2 2 9 7 3
 0 1 5 − 2 3
 
Using the Gaussian–elimination method, we obtain the following sequence of equivalent
augmented matrices:

 1 0 − 3 8 1 1 0 − 3 8 1  1 0 − 3 8 1
  R →R −2R   R ↔R  
2 2 9 7 3 331→  0 2 15 − 9 1  3 
2 →  0 1 5 − 2 3
 0 1 5 − 2 3 0 1 5 − 2 3   0 2 15 − 9 1 
    

BIT MATHEMATICS PROGRAM

75
 MATRIX THEORY

1 0 − 3 8 1  1 0 − 3 8 1 
R →R −2R   R → − 1R  
331→  0 1 5 − 2 3      →  0 1 5 − 2 3 
3 5 3
0 0 5 − 5 − 5   0 0 1 − 1 − 1
  
Since the rank |3 = rank = 3 < 4 , then the system has infinitely many solutions.

% − 3%0 + 8%h = 1
Therefore, the above equation is reduced to

% + 5%0 − 2%h = 3
%0 − %h = −1

⟹ %0 − %h = −1 ⟹ %0 = −1 + %h . Put %h = ,  ∈ ℝ
⟹ %0 = −1 + 
% + 5%0 − 2%h = 3 ⟹ % = 3 + 2%h − 5%0 ⟹ % = 8 − 3
% − 3%0 + 8%h = 1 ⟹ % = 1 + 3%0 − 8%h = −2 − 5
Hence the solution set of the system is m−2 − 5, 8 − 3, −1 + , :  ∈ ℝn
2.8.1.2 GAUSS- JORDAN ELIMINATION
Step 1: Form the augmented matrix [ A b]

Step 2: Transform [ A b] to reduced row echelon form [C d ] using row operations.

Step 3: Solve the system CX = d by back substitution.

% + 3% + 4%0 + %h = 2
Example 2.62: Solve the following system by Gauss-Jordan elimination method.

−2% − 7% − 8%0 + 2%h = 5

3% + % + 13%0 + 4%h = 4

 1 3 4 1 2
 
Solution: The augmented matrix of the system is  − 2 − 7 − 8 2 5 
 3 9 13 4 4 

Using the Gauss–Jordan elimination method, we obtain the following sequence of equivalent
augmented matrices:
 1 3 4 1 2 1 3 4 1 2  1 3 4 1 2 
  R →R +2 R   R →−R  
 − 2 − 7 − 8 2 5 2 2 1  0 − 1 0 4 9  2  
2→  0 1 0 4 − 9
 3 9 13 4 4  R → R −3 R1    0 0 1 1 − 2
 3 3 →  0 0 1 1 − 2  

BIT MATHEMATICS PROGRAM

76
 LINEAR ALGEBRA

 1 0 4 − 11 29   1 0 0 − 15 37 
R →R −3R   R →R −4R  
11 2→ 0
 1 0 4 − 9 11 2→ 0
 1 0 4 − 9
0 0 1 1 − 2 0 0 1 1 − 2 
  
Since the rank |3 = rank = 3 < 4 , then the system has infinitely many solutions.

% − 15%h = 1
Therefore, the above equation is reduced to

% + 4%h = −9
%0 + %h = −2

Put %h = ,  ∈ ℝ, %0 + %h = −2 ⟹ %0 = −2 + %h = −2 + 
% + 4%h = −9 ⟹ % = −9 − 4
% − 15%h = 1 ⟹ % = 1 + 15
Hence the solution set of the system is m1 + 15, −9 − 4, −2 + , :  ∈ ℝn

% + 3% − 2%0 + 2%o = 0
Example 2.63: Solve by Gauss-Jordan elimination

2% + 6% − 5%0 − 2%h + 4%o − 3%e = −1

5%0 + 10%h + 15%e = 5
2% + 6% + 8%h + 4%o + 18%e = 10
Solution: The augmented matrix for the system is
1 3 −2 0 2 0 0 
 
2 6 − 5 − 2 4 − 3 − 1
0 0 5 10 0 15 5 
 
2 6 0 8 4 18 10 

Using the Gauss–Jordan elimination method, we obtain the following sequence of equivalent
augmented matrices:
1 3 −2 0 2 0 0  1 3 −2 0 2 0 0 
   
2 6 − 5 − 2 4 − 3 − 1 R → R − 2 R 0 0 − 1 − 2 0 − 3 − 1
2 2 1
0 0 5 10 0 15 5  0 0 5 10 0 15 5 
  R4 → R4 − 2 R1  
2
 6 0 8 4 18 10    →  0 0 4 8 0 18 10 

BIT MATHEMATICS PROGRAM

77
 MATRIX THEORY

1 3 −2 0 2 0 0  1 3 − 2 0 2 0 0
   
R3 → R3 +5 R2 0 0 − 1 − 2 0 − 3 − 1 R → − R  0 0 1 2 0 3 1
0 2  
2→
0 0 0 0 0 0 0 0 0 0 0 0 0
R → R4 + 4 R2    
  →  0
4
0 0 0 0 6 6  0 0 0 0 0 6 6 
 
1 3 − 2 0 2 0 0 1 3 − 2 0 2 0 0
   
R ↔R 1
0 1 2 0 3 1  R3 → R2  0
3 
4→ 0 6 →
  
0 1 2 0 3 1
0 0 0 0 0 6 6 0 0 0 0 0 1 1
   
0 0 0 0 0 0 0  0 0 0 0 0 0 0 
 
1 3 −2 0 2 0 0  1 3 0 4 2 0 − 4
   
R →R −3R 0 0 1 2 0 0 − 2 R →R +2R 0 0 1 2 0 0 − 2
223→
0 11 2→

0 0 0 0 1 1  0 0 0 0 0 1 1 
   
0 0 0 0 0 0 0  0 0 0 0 0 0 0 
 

% + 3% + 4%h + 2%o = −4
The corresponding system of equation is

%0 + 2%h = −2
%e = 1

%0 + 2%h = −2 ⟹ %0 = −2 − 2%h
From the first and second equation we have,

% + 3% + 4%h + 2%o = −4 ⟹ % = −4 − 3% − 4%h − 2%o

Put ,% =  , %h = , , and %o = i, , ,, i ∈ ℝ . Then the solution set of the system is

m−4 − 3 − 4, − 2i, , −2 − 2,, ,, i: , ,, i ∈ ℝn

% − 2% + 3%0 − 2%h = 0
Example 2.64: solve the homogeneous system

−3% + 6% + %0 =0
−2% + 4% + 4%0 − 2%h = 0
 1 − 2 3 − 2 0
 
Solution: The augmented matrix is  − 3 6 1 0 0 
 − 2 4 4 − 2 0
 
The reduction of the augmented matrix to reduced form is

BIT MATHEMATICS PROGRAM

78
 LINEAR ALGEBRA

 −1 
1 − 2 0 
 1 − 2 3 − 2 0  5 0
  0 0 1 −3
 − 3 6 1 0 0 → 0
 5 
 − 2 4 4 − 2 0 0 0 0 0
  0
 
 

So the solutions are % = 2i +  , % = i, %0 =  , and %h = .

 0
o o

Hence we can write the general solution f in the matrix form

1
2i + 
@ i5 B
f=Y 3 Z , , i ∈ ℝ

5
?  A
Example 2.65: Find the value(s) ℎ so that the system of linear equation
% + ℎ% = 4
3% + 6% = 8

1 ℎ4
a) Has no solution b) has solution

Solution: Augmented matrix is . Q /

3 68

1 ℎ4 1 ℎ 4
. Q / . Q /
3 68 0 6 − 3ℎ −4
R →R −3R
221→

If 6 − 3ℎ = 0, that is, if ℎ = 2, then the system has no solution, because 0 cannot equal −4.
Otherwise, when ℎ ≠ 2, the system has solution.

2.8.1.3 MATRIX INVERSION METHOD

Let a system of n linear equations in n unknowns AX = b , A is invertible, then the system has
unique solution given by inversion method X = A −1b

% + 2% + 3%0 = 11
Example 2.66: Solve the system of linear equation

2% − 4%0 = −6
−% + % + 2%0 = 2

BIT MATHEMATICS PROGRAM

79
 MATRIX THEORY

1 2 3 11
Solution: The coefficient matrix of the system is  = & 0 2 −4*, and 3 = &−6*
−1 1 2 2
|| = 1 Q2 −4Q − Q2 3 Q = 8 + 14 = 22 ≠ 0
1 2 2 −4

and given by inversion method f = J 3.

Therefore, the coefficient matrix is invertible and hence the given system has unique solution,

Inverse of :
2 −4 2 3 2 3
6 = Q Q=8 6 = − Q Q = −1 60 = Q Q = −14
1 2 1 2 2 −4
0 −4 1 3 1 3
6 = − Q Q=4 6 = Q Q=5 60 = −Q Q=4
−1 2 −1 2 0 −4
0 2 1 2 1 2
60 = Q Q=2 60 = − Q Q = −3 600 =Q Q=2
−1 1 −1 1 0 2
The cofactor matrix  of  is
8 4 2
 = & −1 5 −3*
−14 4 2
The adjoint matrix of  is
8 −1 −14
5  =  K = &4 5 4 *
2 −3 2
8 −1 −14
Therefore, the inverse of  is  = |p| = &4 5 4 *
J `a p 

2 −3 2

8 −1 −14 11 66
Thus, the solution of the system is

f =  3 =  &4 5 4 * &−6* =  &22*

J  

2 −3 2 2 44

⟹ % = 3 , % = 1, and %0 = 2

2.8.1.4 CRAMER’S RULE

If f = 3 is a system of  linear equations in  unknowns such that || ≠ 0, then the system

| | | | |0 | | |
has a unique solution. This solution is

% = , % = , %0 = , ⋯ , % =
|| || || ||
Where  is the matrix obtained by replacing the entries in the  − ℎ column of  by the
3
3
entries in the matrix 3 =   
⋮
3

BIT MATHEMATICS PROGRAM

80
 LINEAR ALGEBRA

% + 2% + 3%0 = 0
Example 2.67: Use Cramer’s rule to solve

2% − 4%0 = 2
−% + % + 2%0 = 0
1 2 3 0
2 −4 2 3
Solution:  = & 0 2 −4* , 3 = &2*,|| = 1 Q Q−Q Q = 8 + 14 = 22 ≠ 0
1 2 2 −4
−1 1 2 0
0 2 3
2 3
 = &2 2 −4* ⟹ | | = −2 Q Q = −2
1 2
0 1 2
1 0 3
1 3
 = & 0 2 −4* ⟹ | | = 2 Q Q = 10
−1 2
−1 0 2
1 2 0
1 2
0 = & 0 2 2* ⟹ | | = −2 Q Q = −6
−1 1
−1 1 0

% = =− , % = = , %0 = − =−
J  _ o e 0
Therefore,

If  is an  ×  matrix, then the following statements are equivalent.

Theorem: Equivalent statements

 is invertible.
f = 0 has only the trivial solution.
a)

The reduced roe echelon form of  is - .

b)

f = 3 is consistent for every  × 1 matrix 3.

c)

f = 3 has exactly one solution for every  × 1 matrix 3.

d)

det ≠ 0.
e)
f)
Exercise 2.10:

2% − 4%h = −10 % − 2% + %0 = 6
1. Solve the system of linear equations using the Gaussian–elimination method.
a) b)
3% + 3%0 =0 2% + % − 3%0 = −3
%0 + 4%h = −1 3% + 2% + %0 = 10
−3% + 2% + 3%0 + %h = 5
c) % − 2% + %0 + %h + %o = 1 d) % + 2% + 4%0 + 5%h = 1

BIT MATHEMATICS PROGRAM

81
 MATRIX THEORY

−2% + 5% − 2%0 + %h − 2%o = 4 2% + 4% + 5%0 + 4%h = 3

3% − 6% + 4%0 + 3%h + 3%o = 5 4% + 5% + 4%0 + 2%h = 1
2% − 4% + 2%0 + 3%h + 2%o = 6 −% − 2% − 3%0 − 3%h = 4

e) % + 2% + 4%0 + 5%h = 1 f) % − 3% + 2%h + 2%o = 0

2% + 4% + 5%0 + 4%h = 3 −2% + 6% + %0 + 2%h − 5%o = 0
4% + 5% + 4%0 + 2%h = 1 3% − 9% − %0 + 7%o = 0
−% − 2% − 3%0 − 3%h = 4 −3% + 9% + 2%0 + 6%h − 8%o = 0

% − 2% = 4 % + 5q + r = 1
2. Solve by Cramer’s rule, where applies.
a) b)
2% + 3% = 1 % + q + 2r = 3
% + 5q + 2r = 1
3% − % + %0 = 4 % + 2% − 5%0 = 4
−% + 5% + %0 = −2 % − 4%0 = 3
c) d)

−% − 2% + 5%0 = 1
3. Determine whether the statement is true or false, and justify your answer.
a) If  is 4 × 4 matrix, then |3| = 3||.
b) If  and  are square matrix of the same size such that || = ||, then| + | = 2||
c) If  and  are square matrix of the same size and  is invertible, then |J | = ||
d) A square matrix  is invertible if and only if || = 0
e) If  is a square matrix and the linear system % = 0 has multiple solutions for %,
then|| = 0. .

BIT MATHEMATICS PROGRAM

82
 LINEAR ALGEBRA

UNIT SUMMERY
The arrangement of
 numbers in
rows and  columns is called a matrix of order
× .
A matrix of order  %  is a square matrix.
The elements  ,  , … ,  of a square matrix of order  %  are the main diagonal
elements.
A square matrix in which all the non-diagonal elements are zero is called a diagonal matrix.
Two matrices of the same order are equal if their corresponding entries are equal.
Addition of matrices is commutative and associative.
There is an identity for addition of matrices.

 exists if the rows of A and the columns of B be the same length.

There is an inverse for addition of matrices.

Matrix multiplication is associative but not commutative.

called the transpose of , denoted by K and if  is
× ,then K is  ×
.

A matrix obtained from a given matrix A by interchanging the rows and columns is

For matrices  and  of the same order.

  + K = K +  K

K K = 
>K = >K , > a scalar.


If  has order
%  and  has order  % F then K =  K K .



A square matrix  is symmetric if  = K .

A square matrix  is skew-symmetric if K = −.

Any square matrix  of order  can be written as a product of a symmetric and skew-
symmetric matrix.

Determinant of a square matrix is a scalar.

The determinant of a triangular matrix is the product of its main diagonal elements.
If a row or a column of a square matrix is the zero -vector, then its determinant is zero.
For any square matrix , A = A t

If a row or a column of matrix  is multiplied by a constant k, then the determinate of the new
matrix will be also multiplied by >.

BIT MATHEMATICS PROGRAM

83
 MATRIX THEORY

If two rows or two columns of  are identical, then || = 0.

det> = >  det , for any scalar > and any square matrix  of order .

For a square matrix  such that || ≠ 0, J =

adj A
.
|A|
The three types of elementary row operations that can be performed on matrices are:

Ri ↔ R j (Interchange row  and )

o Interchanging two rows

Ri → kRi (Multiply row  by non zero scalar >)

o Multiplying a row by a non – zero scalar

R j → R j + kRi (Add  th row to th row) .

o Adding a multiple of one row to another row

If the coefficient matrix  of system of  equations in  unknowns is non – singular then

f =  has unique solution, f = J .
Matrices enable us to solve system of linear equations.
A system of linear equations has a unique solution, has no solution or has infinitely many
solutions.
A system of linear equations may be homogeneous or non -homogeneous.
The homogeneous system of linear equations has at least trivial solution.
Every non-homogeneous system of linear equations with more variables than the equations
either has no solution or has infinitely many solutions.
We can solve any system of linear equations using Gaussian elimination or Gauss-Jordan
elimination.
If a system of linear equations has equal number of unknowns and equations and if their
coefficient matrix has nonzero determinant, then we can also solve the system using Matrix-
inverse method or Cramer’s rule.

SOLUTION TO EXERCISES ON UNIT TWO

1. a) Impossible because  and  are of different size.

Exercise 2.1:

b) Impossible because  and  are of different size.

BIT MATHEMATICS PROGRAM

84
 LINEAR ALGEBRA
t
2 1 3 −1 5 0 t 5 2
c)  +  = N.
K
/+. /O = . / =. /
0 −1 2 0 2 −1 0 −1

K
2 1 3 −1 2 6 −1 8 K
K K = N. /. /O = . /
0 −1 0 1 4 0 −1 −4
d) is possible and

6 0
= &−1 −1*
8 −4

2 1 K 3 −1 10 0 3 −1
e) 5K − 6 = 5 . / −. /=. /−. /
0 −1 2 0 5 −5 2 0
7 1
=. /
3 −5
f) Impossible

3 −1 2 1 0 1 4 −1 3
g) L + M,  + M = . /+. /=. /
0 1 4 0 1 0 0 2 4

1 3
4 −1 3
L + M = &−1 0* . /
0 2 4
1 4

4 5 15
= &−4 1 −3*
4 7 19

3 6 − 35 −5
h) Impossible
2. . /=. / ⟺ = 6 − 35
6 5 2 + 5 + 3
3 = −5
6 = 2 + 5
5 = +3
⟹ = 25, 3 = −5, 6 = 55, 5 ∈ ℝ
1 0 5 2 −4 −2 −2 −1
3. a) 5 − . / = 3 − . / ⟹ 2 = . /⟹=. /
2 3 6 1 −4 2 −2 1
1 2 7 2
. / = . /
−1 2 −1 4
3
b)

 must be 2 × 2 matrix. If  = . / , then the equation becomes

6 5
1 2 3 7 2 + 26 3 + 25 7 2
. /. / =. /⟹. /=. /
−1 2 6 5 −1 4 26 − 25 − 3 −1 4
⟹ + 26 = 7 ------- (1)
3 + 25 = 2 ------- (2)
26 − = −1------- (3)
25 − 3 = 4 -------- (4)

BIT MATHEMATICS PROGRAM

85
 MATRIX THEORY

From (1) and (3), we have 6 = , = 4

0

From (2), and (4) , we get 5 = , 3 = −1
0

4 −1
Therefore,  = N 0 0 O
 
K 2 1
1 −1 0
N + 3 . /O = &0 5*
1 2 4
3 8
c)

3 6
 must be 2 × 3 matrix .Let  = v w, then this equation becomes
5 \ [
1 −1 0 3 6 3 −3 0
 +3. /=v w+. /
1 2 4 5 \ [ 3 6 12
3 + −3 + 3 6
=v w, and
3 + 5 6 + \ 12 + [
3+ 5+3 2 1
3 + −3 + 3 6 K
v w = &3 − 3 \ + 6 * = &0 5*
3 + 5 6 + \ 12 + [
6 [ + 12 3 8
⟺ = −1, 3 = 3 , 6 = 3 , 5 = −2, \ = −1, and [ = −4
−1 3 3
Thus,  = . /
−2 −1 −4
, 2, ,
4. Find the values of , and  such that  = &  −1 , * is symmetric.
 , ,
 is symmetric if and only if  = 
K

, 2, , ,  
⟹ &  −1 , * = N2, −1 ,  O ⟺ 2, = , and , = 
 , , , , ,
⟹,= ⟹ =  ⟹   = 2 ⟹   − 2 = 0 ⟹  = 0, or  = 2
K Kx
 
0 0 0
For  = 0, , = 0,  = &0 −1 0*
0 0 0
1 2 2
For  = 2 , , = 1,  = &2 −1 1*
2 1 1

1 2 3
Exercise 2.2:
2 −5 2 3
a) Q Q = 8 + 45 = 53 b) V 2 0 0V = −2 Q Q = −28 − 9 = 2
9 4 3 4
−1 3 4

BIT MATHEMATICS PROGRAM

86
 LINEAR ALGEBRA
5 0 11 −5 0
5 0 11 −5
−3 2 13 7 −4 0 1 5
0 1 5 0
c) 0 1 5 0 0 =4 = 4(5) 2 1 9
2 1 9 0
2 1 9 0 0 1 −1 − 4
1 −1 − 4 0
1 −1 − 4 0 0
0 1 5
 2 9 2 1 
= 4(5) 2 1 9 = 20 − 1 +5  = 20 (17 − 15) = 40

 1 − 4 1 − 1 
1 −1 − 4
5 4 −1 − 3
4 −1 − 3 5 4 −1
d) 1 0 2 1
= 20 2 1 −5 1 0 2 = −6 − 35 = −41
−1 1 6 −2
1 6 −2 −1 1 6
2 0 0 −5

1. a) || = 0, because it contains entirely zero row

Exercise 2.3:

b) || = 4126−19 = 2592

0 0 3 0 0
0 3 0 0
−6 0 0 0 0 0 −5 0
0 0 −5 0 0 9
c) 0 0 0 − 5 0 = 6 = 6 (−3) 0 0 9 = 6 (−3) (5) = 810
0 0 0 9 1 0
0 0 0 0 9 1 0 0
1 0 0 0

d) || = 0, because the third row is a scalar multiple of the first row.
0 1 0 0 0

e) || = 264
> 2>
2. a) Q Q = −12 ⟹ >  − 8> = 12 ⟹ >  − 8> + 12 = 0
4 >
⟹ > − 2> − 6 = 0
⟹ > = 2, or > = 6
3) Given || = 6, || = 3
|32K | = 30 ||20 || = 27863 = 3888
6 3
3. Given V5 [ \ V = 8
]  ℎ

5 [ \ 6 3
a) V 6 3 V = − V5 [ \ V = −8
]  ℎ ]  ℎ
4 43 46 3 6 6 3
b) V−25 −2\ −2[ V = 4−210 V5 \ [ V = −80−1 V5 [ \ V = 640
10] 10ℎ 10 ] ℎ  ]  ℎ

BIT MATHEMATICS PROGRAM

87
 MATRIX THEORY

+] 3+ℎ 6+ 3 6 3 6
c) ^ ] ℎ  ^ = − V] ℎ  V = −−1 V5 \ [V
−5 −\ −[ 5 \ [ ] ℎ 

6 3
= − V5 [ \ V = −8
]  ℎ

9 1 5
Exercise 2.4:

a)  = &1 6 0*
5 2 7
6 = 42 6 = 3 60 = −30
6 = −7 6 = 38 60 = 5
60 = −28 60 = −13 600 = 53

42 −7 −28
Cofactor matrix  is  = & 3 38 −13*, and the adjoint of  is
−30 5 53

42 3 −30
5  =  K = & −7 38 5 *
−28 −13 53
2 3−1
 
b) A =  0 − 42
 1 −1 5 
6 = −18 6 = −14 60 = 2
 

6 = 2 6 = 11 60 = −4
60 = 4 60 = 5 600 = −8
−18 2 4
Cofactor matrix  is  = &−14 11 5 *, and the adjoint of  is
2 −4 −8

−18 −14 2
5  =  = & 2
K
11 −4*
4 5 −8
3 6 −9
c)  = & 0 0 −2*
−2 1 5

6 = 2 6 = −39 60 = −12
6 = 2 6 = −3 60 = 6
60 = 0 60 = −15 600 = 0

BIT MATHEMATICS PROGRAM

88
 2 2 0
LINEAR ALGEBRA

Cofactor matrix  is  = &−39 −3 −15*, and the adjoint of  is

−12 6 0
2 −39 −12
5  =  K = &2 −3 6 *
0 −15 0
0 3 1
d)  = &1 1 2*
3 2 4
6 = 0 6 = −10 60 = 5
6 = 2 6 = −3 60 = 1
60 = −1 60 = 9 600 = −3

0 2 −1
Cofactor matrix  is  = &−10 −3 9 *, and the adjoint of  is
5 1 −3

0 −10 5
5  =  = & 2
K
−3 1*
−1 9 −3

Exercise 2.6:
a) The matrix is in reduced row echelon form.
b) The matrix is in reduced row echelon form.
c) This matrix is not in row-reduced form. Condition 3 is violated: The leading 1 in row 2 lies to
the left of leading one of row 1.
d) The matrix is in reduced row echelon form.
e) The matrix is in reduced row echelon form.
f) This matrix is not in row-reduced form. Condition 1 and 4 are violated: Row 1 consists of all
zeros and does not lie below the nonzero rows. Also, column 4 contains a leading 1 in row 3
and a nonzero element above it.
g) This matrix is not in row-reduced form. Condition 2 is violated. The first nonzero entry in
row 3 is a −1, not a 1
h) This matrix is not in row-reduced form. Condition 4 is violated. Column 3 contains a leading
1 in row 3 and a nonzero element above it. But, it is in row-echelon form.

BIT MATHEMATICS PROGRAM

89
 MATRIX THEORY

Exercise 2.7:
a) Reduced row-echelon form is
12 −4 1  1 0 0
1 0 − 7  →  0
 1 0 

 6 0 9   0 0 1 

b) Reduced row-echelon form is

 1 1 −1 3 1 0 0 0
−1 −1 2 1  →  0
 1 0 0 

− 3 −4 −7 1 0 0 1 0
   
− 2 0 1 6 0 0 0 1

c) Reduced row-echelon form is

1 2 3 1 0 − 1
4 5 6  →  0
 1 2 

 7 8 9   0 0 0 

 0 0 − 2 0 7 12  1 2 − 5 3 0 2
   
d) Reduced row-echelon form is  2 4 − 10 6 12 28  →  0 0 1 0 0 1 
 2 4 − 5 6 − 5 − 1  0 0 0 0 1 2
   

rank  = 3 b) rank  = 4 c) rank  = 2 d) rank  = 3

Exercise 2.8:
a)

∼ ∼
Exercise 2.9:
 1 − 3 2 1 0 0 1 − 3 2 1 0 0 1 − 3 2 1 0 0
     
a)  − 2 2 − 2 0 1 0   0 − 4 2 2 1 0 0 2 0 − 2 0 1
 2 − 4 4 0 0 1 0 2 0 − 2 0 1  0 − 4 2 2 1 0
     

∼ ∼ ∼
     −1
     0 −1 
 1 − 3 2 1 0 0   1 − 3 0 3 − 1 − 2  1 0 0 2 
0 1 0 −1 0 1  0 1 0 −1 0 1  0 1 0 −1 0 1 
 2  2   2 
0 0 1 1  0 0 1 1  0 0 1 1 
 −1 1  −1 1   −1 1 

Therefore, the inverse of  is

 2   2   2 

0
−1 − 


@  B
J = Y−1 0  Z
−1  1

? A

BIT MATHEMATICS PROGRAM

90
 LINEAR ALGEBRA

∼ ∼ ∼
 1 2 
1 2 1 0 1 2 1 0 1 2 1 0  1 0 
d)      8 −1  17 17 
  0 − 17 − 8 1  0 1  −1
8 − 1 0 1     17 17  

0 1 8

 17 17 

 
Therefore, the inverse is

 J
= Tz z
U
−
{
z z

Exercise 2.10:
1. a) The reduction of the augmented matrix to row-echelon form is

∼
 2 0 0 − 4 − 10   1 0 0 − 2 − 5
   
 0 3 3 0 0  0 1 1 0 0 
 0 0 1 4 −1   0 0 1 4 − 1
   
− 3 2 3 1 5  0 0 0 1 1 

So the general solution is %h = 1 , %0 = −5 , % = 5 , % = −3

   

1 −2 1 6 1 −2 1 6
b) The reduction of the augmented matrix to row-echelon form is

&2 1 −3V−3* ∼ T0 1 −1^−3{U

3 2 1 10 0 0 1 −0
The solution set contains one solution % , % , %0  = − , − , − 0
{ z {
0 0
c) The reduction of the augmented matrix to row-echelon form is
 1 − 2 1 1 1 1 1 − 2 1 1 1 1
   
 3 − 6 4 3 3 5 ∼ 0 0 1 0 0 2
 − 2 5 − 2 1 − 2 4 0 1 0 3 0 6
   
 2 − 4 2 3 2 6 0 0 0 1 0 4
−17 − %o
   

So the general solution is f = T −6 U , %o ∈ ℝ

2
4
d) The reduction of the augmented matrix to row-echelon form is

∼
 1
1 2 4 5 1 1 2 4 5 
   1
 2 4 5 4 3  0 1 4 6 1 
 4 5 4 2 1 0 0 1 2 3 
 
 − 1 − 2 − 3 − 3 4   0 0 0 0 14 
   
 3
The system is inconsistent, because the last row would require that 0 =
h
0
if there were a
solution. The solution set is empty.

BIT MATHEMATICS PROGRAM

91
 MATRIX THEORY

e) The reduction of the augmented matrix to row-echelon form is

 1 − 3 0 2 2 0 1 − 3 0 2 2 0
   
 3 − 9 − 1 0 7 0 ∼  0 0 0 0 0 0
− 2 6 1 2 − 5 0  0 0 1 6 −1 0
   
−3 9 2 6 − 8 0   0 0 0 0 0 0
   

% = 3i − 2, − 2 , % = i , %0 = −6, +  , %h = ,, and %o = , where i, ,,  ∈ ℝ.

So the general solution is

1 −2 4
2. a)  = . / ,3 = . /
2 3 1
4 −2 | |
⟹ || = 7 ≠ 0.  = . / ,  = 14
1 3
1 4 | |
 = . / ,  = −7
2 1
⟹ % = |p|| = = 2 , and % = |p|x = − = −1
|p | h |p | z
z z
1 5 1 1
a)  = &1 1 2* , 3 = &3*
1 5 2 1
1 5 1 1 5 1 1 5 1
|| = V1 1 2V = −4,  = &3 | |
1 2*,  = V3 1 2V = −14
1 5 2 1 5 2 1 5 2
1 1 1 1 1 1
 = &1 3 2* , | | = V1 3 2V = 2
1 1 2 1 1 2
1 5 1 1 5 1
0 = &1 1 3* , |0 | = V1 1 3V = 0
1 5 1 1 5 1
% = =  , % = = −  , %0 = =0
|p| | z |px |  |p} |
|p| |p| |p|

1 2 −5 4
b) Cramer's rule does not apply

c)  = & 0 1 −4* , 3 = &3*

−1 −2 5 1
1 2 −5
|| = V 0 1 −4V = 5 − 8 − −8 + 5 = 0. Cramer's rule does not apply
−1 −2 5

3. a) False b) False c) True d) False e) True

BIT MATHEMATICS PROGRAM

92
 LINEAR ALGEBRA

CHEK-LIST
For each of the following, put "
" mark if your answer is "Yes" and Put "X" if your answer
is "NO" in the box provided after each question.

Can you define the notion of matrices?

Can you identify different types of matrices?

Can you compute addition, scalar multiplication and multiplication of matrices?

Can you find transpose of a matrix?

Can you explain symmetric and skew-symmetric matrices?

Can you state the three elementary row operations on matrices?

Can you change a given matrix into its row echelon form?

Can you change a given matrix into its reduced row echelon form?

Can you find the rank of a matrix?

Can you find the inverse of a matrix using elementary row operations?

Can you find the determinant of a given square matrix?

Can you explain properties of determinants?

Can you explain the procedure used to find the adjoint of a square matrix?

Can you find the inverse of a matrix using its adjoint and determinant?

Can you explain what is meant by a system of linear equations?

Can you write the matrix equation form of a given linear system?

Can you write the matrix equation form of a given linear system?

Can you reduce the augmented matrix of a linear system in to echelon form

Can you solve a linear system using elimination method?

Can you solve a linear system using Gauss elimination method?

 Have you completed all the boxes with (
)? If not, please, go back and review
the unit once more, before you go to the next unit

BIT MATHEMATICS PROGRAM

93