347

Matrices 347

8.2.1 Definition .
A rectangular arrangement of numbers (which may be real or complex numbers) in rows
and columns, is called a matrix. This arrangement is enclosed by small ( ) or big [ ] brackets.
The numbers are called the elements of the matrix or entries in the matrix. A matrix is
represented by capital letters A, B, C etc. and its elements by small letters a,b,c,x,y etc. The
following are some examples of matrices:
a 
1 4  2  i 3 2   
A , B  1     , C  [1, 4 , 9] , D   g , E  [l]
 2 3   3 i 5   h 
8.2.2 Order of a Matrix.
A matrix having m rows and n columns is called a matrix of order m×n or simply m×n
matrix (read as 'an m by n matrix). A matrix A of order m×n is usually written in the following
manner
 a11 a12 a13 ...a1 j ...a1n 
a 
 21 a 22 a 23 ...a 2 j ...a 2 n 
 ..... ..... ..... ..... .....  i  1, 2,..... m
A  or A  [a ij ]m n , where
 a i1 a i2 a i3 ...a ij ...a in  j  1, 2,..... n
 ..... ..... ..... ..... ..... 
 
a m 1 a m 2 a m 3 ...a mj ...a mn 
Here aij denotes the element of ith row and jth column. Example : order of matrix
3  1 5 
6 2  7  is 2×3
 
Note :  A matrix of order m×n contains mn elements. Every row of such a matrix
contains n elements and every column contains m elements.
8.2.3 Equality of Matrices .
Two matrix A and B are said to be equal matrix if they are of same order and their
1 6 3  a a 2 a 3 
corresponding elements are equal Example: If A    and B   1  are equal
5 2 1  b1 b 2 b 3 
matrices.
Then a1  1, a 2  6, a 3  3, b1  5, b 2  2, b 3  1
8.2.4 Types of Matrices.
(1) Row matrix : A matrix is said to be a row matrix or row vector if it has only one row and
any number of columns. Example : [5 0 3] is a row matrix of order 1× 3 and [2] is a row
matrix of order 1×1.
 348 Matrices
(2) Column matrix : A matrix is said to be a column matrix or column vector if it has only
 2
one column and any number of rows. Example :  3  is a column matrix of order 3×1 and [2] is a
 6 
column matrix of order 1×1. Observe that [2] is both a row matrix as well as a column matrix.
(3) Singleton matrix : If in a matrix there is only one element then it is called singleton
matrix.
Thus, A  [aij]m n is a singleton matrix if m  n  1 Example : [2], [3], [a], [–3] are singleton
matrices.
(4) Null or zero matrix : If in a matrix all the elements are zero then it is called a zero matrix
and it is generally denoted by O. Thus A  [aij ]mn is a zero matrix if aij  0 for all i and j.

0 0  0 0 0 
Example : [0 ],  ,  , [0 0 ] are all zero matrices, but of different orders.
0 0  0 0 0 
(5) Square matrix : If number of rows and number of columns in a matrix are equal, then it is
a11 a12 a13 
called a square matrix. Thus A  [aij ]mn is a square matrix if m  n . Example : a 21 a 22 a 23  is a
a 31 a 32 a 33 
square matrix of order 3×3
(i) If m  n then matrix is called a rectangular matrix.
(ii) The elements of a square matrix A for which i  j, i.e . a11 , a 22 , a 33 ,....ann are called diagonal
elements and the line joining these elements is called the principal diagonal or leading diagonal
of matrix A.
(iii) Trace of a matrix : The sum of diagonal elements of a square matrix. A is called the
n
trace of matrix A , which is denoted by tr A. tr A   a ii  a11  a 22  ...a nn
i1

Properties of trace of a matrix : Let A  [aii]nn and B  [bij ]nn and  be a scalar
(i) tr(A)   tr( A) (ii) tr( A  B)  tr( A)  tr (B) (iii) tr( AB)  tr(BA)
(iv) tr ( A)  tr ( A' ) or tr ( A T ) (v) tr (In )  n (vi) tr (0)= 0
(vii) tr ( AB )  tr A . tr B
(6) Diagonal matrix : If all elements except the principal diagonal in a square matrix are
zero, it is called a diagonal matrix. Thus a square matrix A  [aij ] is a diagonal matrix if
aij  0, when i  j .

2 0 0 
Example : 0 3 0  is a diagonal matrix of order 3×3, which can be denoted by diag [2, 3,
0 0 4 
4]
Note :  No element of principal diagonal in a diagonal matrix is zero.
 Matrices 349

 Number of zeros in a diagonal matrix is given by n 2  n where n is the order of the

matrix.
 A diagonal matrix of order n  n having d 1 , d 2 ,....., d n as diagonal elements is denoted
by diag [d 1 , d 2 ,..., d n ] .
(7) Identity matrix : A square matrix in which elements in the main diagonal are all '1' and
rest are all zero is called an identity matrix or unit matrix. Thus, the square matrix A  [aij ] is an
1, if i  j
identity matrix, if a ij  
0, if i  j
We denote the identity matrix of order n by In .
 1 0 0
1 0  
Example : [1],   ,  0 1 0  are identity matrices of order 1, 2 and 3 respectively.
0 1   0 0 1 
 
(8) Scalar matrix : A square matrix whose all non diagonal elements are zero and diagonal
elements are equal is called a scalar matrix. Thus, if A  [aij ] is a square matrix and
 , if i  j
aij   , then A is a scalar matrix.
0, if i  j
5 0 0 
1 0   
Example : [2],  , 0 5 0  are scalar matrices of order 1, 2 and 3 respectively.
 0 1  0 0 5 
 
Note :  Unit matrix and null square matrices are also scalar matrices.
(9) Triangular Matrix : A square matrix [a ij ] is said to be triangular matrix if each element
above or below the principal diagonal is zero. It is of two types
(i) Upper Triangular matrix : A square matrix [a ij ] is called the upper triangular matrix, if
aij  0 when i  j .

3 1 2 
Example : 0 4 3  is an upper triangular matrix of order 3×3.
0 0 6 

(ii) Lower Triangular matrix : A square matrix [a ij ] is called the lower triangular matrix, if
aij  0 when i< j.

1 0 0 
Example :  2 3 0  is a lower triangular matrix of order 3×3.
4 5 2 

n(n  1)
Note :  Minimum number of zeros in a triangular matrix is given by where n is
2
order of matrix.
 Diagonal matrix is both upper and lower triangular.
 350 Matrices
 A triangular matrix a  [aij ]nn is called strictly triangular if aij  0 for 1  i  n

Example: 1 A square matrix A  [aij ] in which aij  0 for i  j and a ij  k (constant) for i  j is called a

(a) Unit matrix (b) Scalar matrix (c) Null matrix (d) Diagonal matrix

Solution: (b) When a ij  0 for i  j and a ij is constant for i  j then the matrix [aij ]nn is called a scalar matrix
Example: 2 If A, B are square matrix of order 3, A is non singular and AB  0, then B is a
(a) Null matrix (b) Singular matrix (c) Unit matrix (d) Non singular matrix
Solution: (a) AB = 0 when B is null matrix.
2 5  7
 
Example: 3 The matrix 0 3 11  is known as
0 0 9 

(a) Symmetric matrix (b) Diagonal matrix (c) Upper triangular matrix (d)Skew symmetric matrix
Solution: (c) We know that if all the elements below the diagonal in a matrix are zero, then it is an upper
triangular matrix.
Example: 4 In an upper triangular matrix n×n, minimum number of zeros is [Rajasthan PET 1999]
n(n  1) n(n  1) 2 n(n  1)
(a) (b) (c) (d) None of these
2 2 2
Solution: (a) As we know a square matrix A  [a ij ] is called an upper triangular matrix if a ij  0 for all i>j

a11 a12 a13 a14 .... a1(n  2) a1(n 1) a1n 

 
0 a22 a23 a24 .... a2(n  2) a2(n 1) a2 n 
0 0 a33 a34 .... a3(n  2) a3(n 1) a3 n 
 
A  
0 0 0 a44 .... a4 (n  2) a4 (n 1) a4 n 
. Number of zeros
        
 
        
0 0 0 0 .... 0 a(n 1)(n 1) a(n 1)n 

 0 0 0 0 .... 0 0 ann 
(n  1)n
= (n  1)  (n  2)  .....  2  1 
2
Example: 5 If A  [a ij ] is a scalar matrix then trace of A is

(a)  a
i j
ij (b) ai
ij (c) a
j
ij (d) ai
ii

n
Solution: (d) The trace of A  a
i1
ii  Sum of diagonal elements.

8.2.5 Addition and Subtraction of Matrices.

If A  [aij ]mn and B  [b ij ]m n are two matrices of the same order then their sum A+B is a
matrix whose each element is the sum of corresponding elements. i.e. A  B  [aij  b ij ]mn
5 2 1 5   5  1 2  5  6 7 
Example : If A   1 3  and B   2 2  , then A  B   1  2 3  2   3 5 
  
4 1  3 3  4  3 1  3  7 4 
Similarly, their subtraction A  B is defined as A  B  [aij  b ij ]mn
 Matrices 351

 5  1 2  5  4  3
i.e. in above example A  B  1  2 3  2     1 1 
4  3 1  3   1  2 

Note :  Matrix addition and subtraction can be possible only when matrices are of the
same order.
Properties of matrix addition : If A, B and C are matrices of same order, then
(i) A  B  B  A (Commutative law)
(ii) ( A  B)  C  A  (B  C) (Associative law)
(iii) A  O  O  A  A, where O is zero matrix which is additive identity of the matrix.
(iv) A  ( A)  0  ( A)  A , where ( A) is obtained by changing the sign of every element of
A, which is additive inverse of the matrix.
A  B  A  C
(v)   B  C (Cancellation law)
B  A  C  A
8.2.6 Scalar Multiplication of Matrices.
Let A  [aij ]mn be a matrix and k be a number, then the matrix which is obtained by
multiplying every element of A by k is called scalar multiplication of A by k and it is
denoted by kA.
2 4 10 20 
Thus, if A  [aij ]mn , then kA  Ak  [kaij ]mn . Example : If A   3 
1  , then 5 A  15 5 
4 6   20 30 
Properties of scalar multiplication:
If A, B are matrices of the same order and ,  are any two scalars then
(i) ( A  B)  A  B (ii) (   ) A  A  A
(iii) (A)  (A)  (A) (iv) (A)  (A)   ( A)
Note :  All the laws of ordinary algebra hold for the addition or subtraction of matrices
and their multiplication by scalars.
8.2.7 Multiplication of Matrices.
Two matrices A and B are conformable for the product AB if the number of columns in A
(pre-multiplier) is same as the number of rows in B (post multiplier).Thus, if A  [aij ]mn and
B  [b ij ]n p are two matrices of order m×n and n  p respectively, then their product AB is of order
n
m  p and is defined as ( AB )ij  a
r 1
ir b rj

 b1 j 
b 
 [ai1 ai 2 ...ain ] 2 j  = (ith row of A)(jth column of B) .....(i), where i=1, 2, ..., m and
 
b nj 
j=1, 2, ...p
Now we define the product of a row matrix and a column matrix.
 352 Matrices

b 1 
 
Let A  a1 a 2 ....a n  be a row matrix and B  b 2  be a column matrix.
  
b n 
Then AB  a1 b1  a 2 b 2  ....  an b n  …(ii). Thus, from (i),
( AB)ij  Sum of the product of elements of ith row of A with the corresponding elements of jth
column of B.
Properties of matrix multiplication
If A,B and C are three matrices such that their product is defined, then
(i) AB  BA (Generally not commutative)
(ii) ( AB)C  A(BC ) (Associative Law)
(iii) IA  A  AI , where I is identity matrix for matrix multiplication
(iv) A(B  C)  AB  AC (Distributive law)
(v) If AB  AC 
 BC (Cancellation law is not applicable)
(vi) If AB= 0 It does not mean that A= 0 or B = 0, again product of two non
zero matrix may be a zero matrix.
Note :  If A and B are two matrices such that AB exists, then BA may or may not exist.
 The multiplication of two triangular matrices is a triangular matrix.
 The multiplication of two diagonal matrices is also a diagonal matrix and
diag (a1 , a 2 ,....a n )  diag (b1 , b 2 ,....b n )  diag (a1 b1 , a 2 b 2 ,....an b n )
 The multiplication of two scalar matrices is also a scalar matrix.
 If A and B are two matrices of the same order, then
(i) ( A  B)2  A 2  B 2  AB  BA (ii) ( A  B 2 )  A 2  B 2  AB  BA
(iii) ( A  B)(A  B)  A 2  B 2  AB  BA (iv) ( A  B)(A  B)  A 2  B 2  AB  BA
(v) A( B)  ( A)B  ( AB)
8.2.8 Positive Integral Powers of A Matrix.
The positive integral powers of a matrix A are defined only when A is a square matrix. Also
then A 2  A. A , A 3  A. A. A  A 2 A . Also for any positive integers m ,n.
(i) A m A n  A m n (ii) ( A m )n  A mn  ( A n )m
(iii) I n  I, I m  I (iv) A 0  In where A is a square matrix of
order n.
8.2.9 Matrix Polynomial .
Let f (x )  a0 x n  a1 x n 1  a 2 x n  2  ...  an 1 x  an be a polynomial and let A be a square matrix of
order n. Then f ( A)  a0 A n  a1 A n 1  a 2 A n  2  ...  an 1 A  an In is called a matrix polynomial.
Example : If f (x )  x 2  3 x  2 is a polynomial and A is a square matrix, then A 2  3 A  2 I is a
matrix polynomial.

 cos  sin  
Example: 6 If A   , then A 
2
[Rajasthan PET 2001]
 sin  cos  
 Matrices 353

cos  sin   cos   sin    cos  sin    cos  sin  

(a)   (b)   (c)   (d)  
 sin  cos    sin  cos    sin  cos     sin   cos  

 cos  sin    cos  sin    cos 2 sin 2 

Solution: (c) Since A 2  A. A      
 sin  cos    sin  cos    sin 2 cos 2 
a b   
If A    and A  
2
 
Example: 7 then [AIEEE 2003]
b a  
(a)   a 2  b 2 ,   ab (b)   a 2  b 2 ,   2ab (c)   a2  b 2 ,   a2  b 2 (d)   2ab,   a 2  b 2

   a b  a b  a 2  b 2 2 ab 
A2       . On comparing, we get,   a  b ,   2ab
2 2
Solution: (b) 
    b a  b a   2 ab a 2  b 2 

 i 0
Example: 8 If A   , n  N , then A equals
4n
[AMU 1992]
0 i 
1 0   i 0 0 i  0 0 
(a)   (b)   (c)   (d)  
0 1  0 i   i 0 0 0 
 i 0   i 0  1 0  1 0  1 0  1 0  1 0 
Solution: (a) A2     , A4  A2. A2       I ; (A )  I  I  
4 n n

0 i  0 i   0  1  0  1  0  1 0 1  0 1 
1 1  a 1 
Example: 9 If A   , B    and ( A  B)  A  B then value of a and b are
2 2 2
[Kurukshetra CEE 2002]
 2  1 b  1
(a) a  4 , b  1 (b) a  1, b  4 (c) a  0, b  4 (d) a  2 , b  4

Solution: (b) We have ( A  B)2  A 2  B 2  A.B  BA

 a  b 2 a  2 a  1 
∵ AB  BA  0     0
2a  b 3  b  2  b  1
2a  2  b a  1
   0 . On comparing, we get, a  1  0  a  1 and 4  b  0  b  4
 2a  2 4 b 

a h g x 
   
Example: 10 The order of [ x y z ]  h b f  y  is [EAMCET 1994]
 g f c   z 

(a) 3×1 (b) 1×1 (c) 1×3 (d) 3×3

Solution: (b) Order will be (1  3)(3  3)(3  1)  (1  1)
cos   sin  0
 
Example: 11 Let F( )   sin  cos  0  . Then F ( ).F ( ' ) is equal to
 0 0 1 
(a) F (' ) (b) F ( /  ' ) (c) F (   ' ) (d) F (   ' )
cos   sin  0 cos    sin   0 
   
Solution: (c) We have F( )   sin  cos  0  , F( )   sin   cos   0 
 0 0 1   0 0 1 

cos   sin  0  cos    sin   0  cos(   )  sin(   ) 0 

    
F ( ).F ( )   sin  cos  0   sin   cos   0    sin(   ) cos(   ) 0   F (   )
 0 0 1   0 0 1   0 0 1 

1 1 0 
 
Example: 12 For the matrix A  1 2 1  , which of the following is correct
 2 1 0 

(a) A 3  3 A 2  I  0 (b) A 3  3 A 2  I  0 (c) A 3  2 A 2  I  0 (d) A 3  A 2  I  0

 354 Matrices
1 1 0  1 1 0   2 3 1 2 3 1 1 1 0   7 9 3
     , 3      
Solution: (b) A  A. A  1 2 1  1 2 1   5 6
2
2  A  A. A  5 6
2
2 1 2 1   15 19 6
 2 1 0   2 1 0  3 4 1  3 4 1   2 1 0   9 12 4 

7 9 3  6 9 3  1 0 0 
     
A  3 . A  15 19
3 2
6   15 18 6   0 1 0   I  A 3  3 A 2  I  0
 9 12 4   9 12 3  0 0 1 

 0  1 0 
If A   , B    , then the value of  for which A  B is
2
Example: 13 [IIT Screening 2003]
 1 1  5 1 
(a) 1 (b) –1 (c) 4 (d) No real values
 0   0   2
0
A2     ∵ A  B (given)
2
Solution: (d) 
1 1  1 1    1 1 

 2 0  1 0 
     1 and   1  5 . Clearly no real value of 
2
Then 
  1 1  5 1 

8.2.10 Transpose of a Matrix.

The matrix obtained from a given matrix A by changing its rows into columns or columns
into rows is called transpose of Matrix A and is denoted by A T or A  .
From the definition it is obvious that if order of A is m×n, then order of A T is n×m
 a1 b1 
a a3 
is a 2 b 2 
a2
Example : Transpose of matrix  1
b1 b2 b 3  23
a 3 b 3  3 2
Properties of transpose : Let A and B be two matrices then
(i) ( AT )T  A
(ii) ( A  B)T  AT  BT , A and B being of the same order
(iii) (kA)T  kAT , k be any scalar (real or complex)
(iv) ( AB)T  BT AT , A and B being conformable for the product AB
(v) ( A1 A 2 A 3 ..... An 1 An )T  An An 1 ....... A 3 A 2 A1
T T T T T

(vi) I T  I
8.2.11 Determinant of a Matrix .
 a11 a12 a13 
If A  a 21 a 22 a 23  be a square matrix, then its determinant, denoted by |A| or Det (A) is
a 31 a 32 a 33 
defined as
a11 a12 a13
| A |  a 21 a 22 a 23
a 31 a 32 a 33
Properties of determinant of a matrix
(i) | A | exists  A is square matrix (ii) | AB | | A || B |
(iii) | A T | | A | (iv) | kA |  k n | A |, if A is a square matrix of
order n
(v) If A and B are square matrices of same order then |AB|=|BA|
 Matrices 355

(vi) If A is a skew symmetric matrix of odd order then | A |  0

(vii) If A  diag (a1 , a 2 ,.....a n ) then | A |  a1 a 2 ...a n (viii)| A | n | A n |, n  N .
Example: 14 If A and B are square matrices of same order then
[Pb. CET 1992; Roorkee 1995; MP PET 1990; Rajasthan PET
1992, 94]
(a) ( AB )  A B  (b) ( AB )  B A 
(c) AB  0, if | A |  0 or | B |  0 (d) AB  0, if | A |  I or B  I
Solution: (b) A  [aij ]nn , B  [b jk ]nn , AB  [aij ]nn [b jk ]nn  [cik ]nn , where c ik  aij b jk
(AB )  [cik ]nn  [cki ]nn  [b kj ]nn [a ji ]nn = B A
1 2  1 3  1 11 
Alternatively, Let A    ,B    ; AB   
 3 4  2 2 0 4  2 2 3 25 
1 3  1 0  1 3   1 3 
( AB )'    …..(i) and B ' A '      …..(ii)
11 25  3 4  2 4  11 25 
From (i) and (ii), ( AB )  B A 
Example: 15 If A,B are 3×2 order matrices and C is a 2×3 order matrix, then which of the following matrices not
defined
[Rajasthan PET 1998]
(a) A t  B (b) B  C t (c) A t  C (d) A t  B t
Solution: (a) Order of A is 3 × 2 and order of B is 3 × 2 and order of A t is 2 × 3 then
 A t  B  [ A t ]23 + [B]32 is not possible because order are not same.
8.2.12 Special Types of Matrices.
(1) Symmetric and skew-symmetric matrix
(i) Symmetric matrix : A square matrix A  [aij] is called symmetric matrix if aij  a ji for all i, j
or A T  A
a h g 
Example :  h b f 
 g f c 
Note :  Every unit matrix and square zero matrix are symmetric matrices.
n(n  1)
 Maximum number of different elements in a symmetric matrix is
2
(ii) Skew-symmetric matrix : A square matrix A  [aij] is called skew- symmetric matrix if
aij  a ji for all i, j
 0 h g

or A   A . Example :   h 0
T
f 
 g  f 0 
Note :  All principal diagonal elements of a skew- symmetric matrix are always zero
because for any diagonal element. aij  aij  aij  0
 Trace of a skew symmetric matrix is always 0.
Properties of symmetric and skew-symmetric matrices:
(i) If A is a square matrix, then A  A T , AA T , A T A are symmetric matrices, while A  A T is skew-
symmetric matrix.
(ii) If A is a symmetric matrix, then  A, KA, A T , A n , A 1 , B T AB are also symmetric matrices,
where n  N , K  R and B is a square matrix of order that of A
 356 Matrices
(iii) If A is a skew-symmetric matrix, then
(a) A 2 n is a symmetric matrix for n  N ,
(b) A 2n 1 is a skew-symmetric matrix for n  N ,
(c) kA is also skew-symmetric matrix, where k  R ,
(d) B T AB is also skew- symmetric matrix where B is a square matrix of order that of A.
(iv) If A, B are two symmetric matrices, then
(a) A  B, AB  BA are also symmetric matrices,
(b) AB  BA is a skew- symmetric matrix,
(c) AB is a symmetric matrix, when AB  BA .
(v) If A,B are two skew-symmetric matrices, then
(a) A  B, AB  BA are skew-symmetric matrices,
(b) AB  BA is a symmetric matrix.
(vi) If A a skew-symmetric matrix and C is a column matrix, then C T AC is a zero matrix.
(vii) Every square matrix A can uniquelly be expressed as sum of a symmetric and skew-
symmetric matrix i.e.
1  1 
A   ( A  A T )   ( A  A T ) .
2  2 
(2) Singular and Non-singular matrix : Any square matrix A is said to be non-singular if
| A |  0, and a square matrix A is said to be singular if | A |  0 . Here | A | (or det(A) or simply det
|A| means corresponding determinant of square matrix A.
2 3 2 3
Example : A    then| A |   10  12  2  A is a non singular matrix.
4 5  4 5
(3) Hermitian and skew-Hermitian matrix : A square matrix A  [aij ] is said to be hermitian
 3 3  4 i 5  2i 
 a b  ic 
matrix if aij  a ji 
i. j i.e. A  A . Example :   , 3  4 i 5  2  i are Hermitian
b  ic d   5  2i  2  i
 2 
matrices.
Note :  If A is a Hermitian matrix then aii  aii  aii is real i, thus every diagonal element
of a Hermitian matrix must be real.
 A Hermitian matrix over the set of real numbers is actually a real symmetric
matrix and a square matrix, A=|aij| is said to be a skew-Hermitian if
aij  a ji. i, j i.e. A   A .
 3i  3  2i  1  i 
 0  2  i 
Example :   ,  3  2i  2i  2  4 i are skew-Hermitian matrices.
 2  i 0   1i
 2  4i 0 
 If A is a skew-Hermitian matrix, then aii  aii  aii  aii  0 i.e. aii must be purely
imaginary or zero.
 A skew-Hermitian matrix over the set of real numbers is actually a real skew-
symmetric matrix.
(4) Orthogonal matrix : A square matrix A is called orthogonal if AA T  I  A T A i.e. if A 1  A T
 cos   sin    cos  sin  
Example : A    is orthogonal because A 1     AT
 sin  cos    sin  cos  
 Matrices 357

In fact every unit matrix is orthogonal.

(5) Idempotent matrix : A square matrix A is called an idempotent matrix if A 2  A .
1 / 2 1 / 2  1 / 4  1 / 4 1 / 4  1 / 4  1 / 2 1 / 2
Example :   is an idempotent matrix, because A 2     A.
1 / 2 1 / 2  1 / 4  1 / 4 1 / 4  1 / 4  1 / 2 1 / 2
1 0  0 0 
Also, A   and B    are idempotent matrices because A  A and B  B .
2 2

 0 0   0 1 
In fact every unit matrix is indempotent.
(6) Involutory matrix : A square matrix A is called an involutory matrix if A 2  I or A 1  A
1 0  1 0 
Example : A    is an involutory matrix because A 2   I
0 1  0 1 
In fact every unit matrix is involutory.
(7) Nilpotent matrix : A square matrix A is called a nilpotent matrix if there exists a
p  N such that A p  0
0 0  0 0 
Example : A    is a nilpotent matrix because A 2   0 (Here P = 2)
1 0  0 0 
(8) Unitary matrix : A square matrix is said to be unitary, if A' A  I since | A  | | A | and
| A ' A | | A '|| A | therefore if A  A=I, we have | A '|| A |  1
Thus the determinant of unitary matrix is of unit modulus. For a matrix to be unitary it
must be non-singular.
Hence A  A  I  A A   I
(9) Periodic matrix : A matrix A will be called a periodic matrix if A k 1  A where k is a
positive integer. If, however k is the least positive integer for which Ak 1  A, then k is said to be
the period of A.
 f (x ) g(x ) dA  f (x ) g (x )
(10) Differentiation of a matrix : If A    then  is a differentiation of
h(x ) l(x )  dx h(x ) l(x ) 
matrix A.
 x 2 sin x  dA  2 x cos x 
Example : If A    then 
2 x 2  dx  2 0 
(11) Submatrix : Let A be m×n matrix, then a matrix obtained by leaving some rows or
2 1 0 
columns or both, of A is called a sub matrix of A. Example : If A'  3 2 2  and 
2 2
 are sub
5 3
 2 5 3 
 2 1 0  1
matrices of matrix A  3 2 2 4 
 2 5 3 1 
(12) Conjugate of a matrix : The matrix obtained from any given matrix A containing complex
number as its elements, on replacing its elements by the corresponding conjugate complex
 358 Matrices

1  2i 2  3i 3  4 i
numbers is called conjugate of A and is denoted by A . Example : A  4  5 i 5  6 i 6  7 i  then
 8 7  8i 7 
 1  2i 2  3i 3  4 i
A  4  5 i 5  6 i 6  7 i
 8 7  8i 7 
Properties of conjugates :
 
(i) A  A (ii)  A  B   A  B
(iii) (A)   A,  being any number (iv) ( AB)  A B , A and B being conformable for
multiplication.
(13) Transpose conjugate of a matrix : The transpose of the conjugate of a matrix A is called
transposed conjugate of A and is denoted by A  . The conjugate of the transpose of A is the same
as the transpose of the conjugate of A i.e. ( A)  ( A )  A .
If A  [aij ]mn then A  [b ji]nm where b ji  a ij i.e. the ( j, i) th element of A   the conjugate of
(i, j) th element of A.
1  2i 2  3i 3  4 i  1  2i 4  5 i 8 
Example : If A  4  5 i 5  6 i 6  7 i  , then A   2  3 i 5  6 i 7  8 i
 
 8 7  8i 7  3  4 i 6  7 i 7 
Properties of transpose conjugate
(i) ( A  )  A (ii) ( A  B)  A  B (iii) (kA)  KA , K being any number (iv)
  
( AB)  B A
0 5  7
 
Example: 16 The matrix  5 0 11  is known as [Karnataka CET 2000]
 7  11 0 
(a) Upper triangular matrix (b) Skew-symmetric matrix (c) Symmetric matrix (d)
Solution: (b) In a skew-symmetric matrix, aij  a ji  i, j  1,2,3 and j  i , aii  aii  each aii =0
Hence the given matrix is skew-symmetric matrix [ A T   A] .
2   4
 
Example: 17 The matrix  1 3 4  is non singular if [Kurukshetra CEE 2002]
 1  2  3 
(a)   2 (b)   2 (c)   3 (d)   3
2   4
 
Solution: (a) The given matrix A   1 3 4  is non singular If |A|  0
 1  2  3 
2  4 1 3 0
 | A|  1 3 4  0  1 3 4  0 [R1  R1  R2 ]
1 2 3 1 2 3
1   3 0 
   R 2  R 2  R3 
 | A |  0 1 1 0 ,  
0    5  3   R3  R3  R1 

 1(3    5 )  0    2  0    2
 Matrices 359

1 2 2 
1 
Example: 18 The matrix A  2 1  2  is [Kurukshetra CEE 2002]
3
 2 2  1 
(a) Orthogonal (b) Involutary (c) Idempotent (d) Nilpotent
1 2 2
1
Solution: (a) Since for given A  2 1  2 . For orthogonal matrix AA T  AT A  I(33)
3
 2 2 1
 1 2 2  1 2  2  9 0 0 
1   1 
 AA T  2 1  2 2 1 2 
 3 0 9 0   3 I . Similarly A A  3 I . Hence A is orthogonal
T
3 
 2 2  1   2  2  1  0 0 9 
 4 x  2
Example: 19 If A    is symmetric, then x = [Karnataka CET 1994]
2 x  3 x  1
(a) 3 (b) 5 (c) 2 (d) 4
 4 2 x  3  4 x  2
Solution: (b) For symmetric matrix, A  A T      2x  3  x  2  x  5
 x  2 x  1  2 x  3 x  1
Example: 20 If A and B are square matrices of order n×n, then ( A  B)2 is equal to [Karnataka CET 1999; Kerala
(Engg.) 2002]

(a) A 2  B 2 (b) A 2  2 AB  B 2 (c) A 2  2 AB  B 2 (d) A 2  AB  BA  B 2

Solution: (d) Given A and B are square matrices of order n×n we know that
( A  B)2  ( A  B)( A  B)  A 2  AB  BA  B 2

cos   sin  
Example: 21 If A   , then which of the following statement is not correct [DCE 2001]
 sin  cos  

(a) A is orthogonal matrix (b) A T is orthogonal matrix (c) Determinant A  1 (d)

Solution: (d) | A |  1  0, therefore A is invertible. Thus (d) is not correct

Example: 22 Matrix A is such that A 2  2 A  I where I is the identity matrix. Then for n  2, A n 

(a) nA  (n  1)I (b) nA  I (c) 2n 1 A  (n  1)I (d) 2 n 1 A  I

Solution: (a) We have, A 2  2 A  I  A 2 . A  (2 A  I)A ; A 3  2 A 2  IA  2 [2 A  I]  IA  A 3  3 A  2 I

Similarly A 4  4 A  3 I and hence An  nA  (n  1)I

 0 0  1
 
Example: 23 Let A   0  1 0  , the only correct statement about the matrix A is [AIEEE 2004]
 1 0 0 

(a) A 2  I (b) A  (1)I , where I is unit matrix

(c) A 1 does not exist (d) A is zero matrix

 0 0  1  0 0  1  1 0 0
    
Solution: (a) A  A. A   0
2
1 0  0 1 0   0 1 0   I . Also, A 1 exists as | A |  1
 1 0 0   1 0 0   0 0 1 

8.2.13 Adjoint of a Square Matrix.

Let A  [a ij ] be a square matrix of order n and let C ij be cofactor of aij in A. Then the
transpose of the matrix of cofactors of elements of A is called the adjoint of A and is denoted by
adj A
360 Matrices

Thus, adj A  [Cij ]T  (adj A)ij  C ji  cofactor of a ji in A.

T
 a11 a12 a13   C11 C12 C13  C11 C 21 C 31 
If A  a 21 a 22 
a 23 , then adj A  C 21 C 22 
C 23   C12 C 22 C 32 ;
a 31 a 32 a 33  C 31 C 32 C 33  C13 C 23 C 33 
Where C ij denotes the cofactor of aij in A.

p q
Example : A   , C 11  s, C 12  r, C 21  q, C 22  p
r s
T
 s  r  s  q
 adj A     
 q p   r p 
Note :  The adjoint of a square matrix of order 2 can be easily obtained by interchanging
the diagonal elements and changing signs of off diagonal elements.
 Matrices 359

Properties of adjoint matrix : If A, B are square matrices of order n and I n is corresponding unit matrix,
then (i) A(adj A) | A | In  (adj A) A (Thus A (adj A) is always a scalar matrix)

(ii) | adj A | | A | n 1 (iii) adj (adj A) | A | n  2 A

(iv) | adj (adj A)| | A | (n 1)

2
(v) adj( A T )  (adj A)T

(vi) adj ( AB )  (adj B)(adj A) (vii) adj( A m )  (adj A)m , m  N

(viii) adj(kA)  k n 1 (adj A), k  R (ix) adj (In )  In

(x) adj (O)  O (xi) A is symmetric  adj A is also symmetric.
(xii) A is diagonal  adj A is also diagonal. (xiii) A is triangular  adj A is also triangular.
(xiv) A is singular  |adj A|= 0
8.2.14 Inverse of a Matrix.
A non-singular square matrix of order n is invertible if there exists a square matrix B of the same order
such that AB  In  BA .
In such a case, we say that the inverse of A is B and we write A 1  B
1
The inverse of A is given by A 1  .adj A
| A|
The necessary and sufficient condition for the existence of the inverse of a square matrix A is that
| A| 0
Properties of inverse matrix:
If A and B are invertible matrices of the same order, then
(i) ( A 1 )1  A (ii) ( A T )1  ( A 1 )T
(iii) ( AB)1  B 1 A 1 (iv) ( A k )1  ( A 1 )k , k  N [In particular ( A 2 )1  ( A 1 )2 ]
1
(v) adj( A 1 )  (adj A)1 (vi) | A 1 |  | A | 1
| A|
(vii) A = diag (a1 a 2 ...a n )  A 1  diag(a11 a 21 ...an1 ) (viii) A is symmetric  A 1 is also symmetric.
(ix) A is diagonal, | A |  0  A 1 is also diagonal. (x) A is scalar matrix  A 1 is also scalar matrix.
(xi) A is triangular, | A |  0  A 1 is also triangular. (xii) Every invertible matrix possesses a unique
inverse.
Note :  (Cancellation law with respect to multiplication)
If A is a non singular matrix i.e., if | A |  0 ,then A 1 exists and AB  AC  A 1 ( AB)  A 1 ( AC)
 ( A 1 A)B  ( A 1 A)C  IB  IC  B  C  AB  AC  B  C | A |  0

4 2 
Example: 24 If A    , then | adj A | is equal to [UPSEAT 2003]
3 4 
(a) 16 (b) 10 (c) 6 (d) None of these

 4 2 
Solution: (b) adj A   
 3 4 
 360 Matrices
4 2
| adj A |   16  6  10
3 4
Example: 25 If 3, – 2 are the Eigen values of non-singular matrix A and |A| = 4 . Then Eigen values of adj (A) are
[Kurukshetra CEE 2002]
(a) 3/4, –1/2 (b) 4/3, –2 (c) 12, –8 (d) –12, 8
adj A
Solution: (b) Since A 1  and if  is Eigen value of A then 1 is Eigen value of A 1 , thus for adj ( A) x  ( A 1 x )| A |
| A|

| A | .1 I
adj(A) corresponding to Eigen value
  3 is = 4/3 and for   2 is = 4 /  2 = –2
3 2 4
  1
Example: 26 If matrix A  1 2  1  and A 1  adj ( A) , then K is [UPSEAT 2002]
K
0 1 1 
(a) 7 (b) –7 (c) 1/7 (d) 11
adj( A) 1
Solution: (d) We know that A 1  . We have A 1  adj( A) i.e. K | A |
| A| K

3 2 4
and K  1 2  1  3(3)  2(1)  4 (1)  9  2  4  11
0 1 1

a b 
Example: 27 The inverse of matrix A    is
c d 
[AMU 2001]
 d b  1  d b  1 1 0   b a 
(a)   (b)   (c)   (d)  
 c a  ad  bc  c a  | A | 0 1  d  c 
a b  d b  1 1  d b 
Solution: (b) Here | A |   ad  bc , adj ( A)    . Hence A   
c d   c a  ad  bc  c a 
1  1 1  4 2 2
   
Example: 28 Let A   2 1  3  and 10 . B   5 0   . If B is the inverse of matrix A, then  is [AIEEE 2004]
1 1 1   1  2 3 

(a) 5 (b) –1 (c) 2 (d) –2

 1 1 1  4 2 2
   
Solution: (a) We have, A   2 1  3  ,  | A |  1(4)  1(5)  1(1)  10 and adj (A)   5 0 5
 1 1 1   1  2 3 

4 2 2
1  
Then A 1  5 0 5
10 
 1  2 3 
According to question, B is the inverse of matrix A. Hence   5
 1 0  K
 
Example: 29 Matrix A   2 1 3  is invertible for [UPSEAT 2002]
 K 0 1 

(a) K  1 (b) K  1 (c) K 0 (d) All real K

1 0 K
Solution: (d) For invertible, | A |  0 i.e., 2 1 3  0
K 0 1
 Matrices 361

 1(1)  K(K)  0 | A |  K 2  1  0 , which is true for all real K .

Example: 30 Let p be a non-singular matrix, [Orissa JEE 2002]
1
1  p  p  .......  p  0
2 n
(0 denotes the null matrix), then p =
(a) p n
(b)  p n
(c)  (1  p  .....  p n ) (d) None of these
Solution: (a) We have, 1  p  p      p  0
2 n

Multiplying both sides by p 1 , p 1  I  Ip  .....  p n 1 I  0. p 1

p 1  I(1  p  ....  p n 1 )  0  p 1   I(1  p  p 2  ....  p n 1 )   ( p n )  p n .
cos   sin  0
 
Example: 31 Let f ( )   sin  cos  0  , where   R ,then [ f ( )]1 is equal to [AMU 2000]
 0 0 1 
(a) f ( ) (b) f ( 1 ) (c) f (2 ) (d) None
cos   sin  0 cos  sin  0
Solution: (a) | f ( )|  sin  cos  0  1 , adj of f ( )   sin  cos  0
0 0 1 0 0 1
cos  sin  0 cos  sin  0
1
[ f ( )]   sin  cos  0 ......(i) and f ( )   sin  cos  0 ......(ii)
0 0 1 0 0 1

From (i) and (ii),  f ( )1  f [ ]

Example: 32 If I is a unit matrix of order 10, then the determinant of I is equal to [Kerala (Engg.) 2002]
(a) 10 (b) 1 (c) 1/10 (d) 9
Solution: (b) Determinant of unit matrix of any order =1.
 2 
If A    and | A |  125 then  =
3
Example: 33 [IIT Screening 2004]
2  
(a)  3 (b)  2 (c) 5 (d) 0
Solution: (a) 125 | A | | A | | A |  5 and | A |    4  5    9     3
3 3 2 2

Example: 34 If |A| denotes the value of the determinant of the square matrix A of order 3, then | 2 A |  [MP PET 1987, 89, 92, 2000]
(a) 8 | A | (b) 8 | A | (c) 2 | A | (d) None of these
Solution: (a) We know that, det. ( A)  (1) det A , where n is order of square matrix
n

If A is square matrix of order 3, Then n  3 . Hence | 2 A |  (2)3 | A |  8 | A | .

 3 1  x 2 
 
Example: 35 For how many value (s) of x in the closed interval [–4, –1] is the matrix  3 1 x  2  singular
 x  3 1 2 
[Karnataka CET 2002]
(a) 2 (b) 0 (c) 3 (d) 1
3 x 1 2
Solution: (d) 3 1 x  2  0
x 3 1 2
0 x x 0 x x
3 1 x2 0 [R1  R1  R2 ] , x 0 x 0 [R2  R2  R3 ]
x 3 1 2 x 3 1 2
0 0 x
x x x 0 [C2  C2  C3 ]
x 3 1 2
 x[( x )  x(x  3)]  0  x (x 2  4 x )  0  x  0,  4
 362 Matrices
Hence only one value of x in closed interval [–4,–1] i.e. x  4
Example: 36 Inverse of diagonal matrix (if it exists) is a
(a) Skew-symmetric matrix (b) Diagonal matrix (c) Non invertible matrix (d) None of these
Solution: (b) Let A  diag (d1, d2 , d3 ......, dn )
As A is invertible, therefore det( A)  0  d1, d2 , d3 ......., dn  0  d i  0 for i = 1, 2, 3…..n
Here, cofactor of each non diagonal entry is 0 and cofactor of aii
1 | A|
 (1)i1 det [diag(d1 , d 2 , d 3 ....., di1 , di1 ....., dn )]  d1 , d 2 , d 3 ..... di1 .di1 ,......, dn  [d1 , d 2 , d 3 ....., di1 , di , di1 ....., dn ] 
di di
1 1 1 1 
A 1  (adj A)  diag  , .......,  , which is a diagonal matrix
| A|  d1 d 2 dn 

8.2.15 Elementary Transformations or Elementary Operations of a Matrix.

The following three operations applied on the rows (columns) of a matrix are called elementary row
(column) transformations
(1) Interchange of any two rows (columns)
If ith row (column) of a matrix is interchanged with the ith row (column), it will be denoted by
R i  R j (C i  C j )
 2 1 3  2 1 3
 
Example : A   1 2 1  , then by applying R 2  R3 , we get B   3 2 4 
 3 2 4   1 2 1 
(2) Multiplying all elements of a row (column) of a matrix by a non-zero scalar
If the elements of ith row (column) are multiplied by a non-zero scalar k, it will be denoted by
Ri  Ri(k ) , [Ci  Ci (k )] or Ri  kR i , [Ci  kC i ]
 3 2  1  3 2  1

If A   0 1 2  , then by applying R 2  3 R 2 we obtain B   0 3
 6 
 1 2  3   1 2  3 
(3) Adding to the elements of a row (column), the corresponding elements of any other row (column)
multiplied by any scalar k. If k times the elements of jth row (column) are added to the corresponding elements
of the ith row (column), it will be denoted by R i  R i  kR j C i  C i  kC j 
 2 1 3 1
If A   1  1 0 2 , then the application of elementary operation R 3  R 3  2R1 gives the matrix
 0 1 3 1 
2 1 3 1

B   1  1 0 2 , If a matrix B is obtained from a matrix A by one or more elementary transformations,
 4 3 9 3 
then A and B are equivalent matrices and we write A ~ B
 1 2 3 4 1 2 3 4
  
Let A   2 1 4 3 , then A ~  1  1 1  1, applying R 2  R 2  (1)R1
 3 1 2 4   3 1 2 4 
1 2 3 1
~  1  1 1  2  , applying C 4  C 4  (1)C 3

 1 1 2 2 
An elementary transformation is called a row transformation or a column transformation according as it
is applied to rows or columns.
 Matrices 363

8.2.16 Elementary Martix.

A matrix obtained from an identity matrix by a single elementary operation (transformation) is called an
 1 3 0  0 0 1   1 0 0
elementary matrix. Example : 0 1 0  0 1 0 , 0 0 1  are elementary matrices obtained from I 3 by
0 0 1  1 0 0  0 1 0 
subjecting it to the elementary transformations R1  R1  3 R 2 , C1  C3 and R 2  R3 respectively.
Theorem 1 : Every elementary row (column) transformation of an m×n matrix (not identity matrix) can be
obtained by pre-multiplication (post- multiplication) with the corresponding elementary matrix obtained from the
identity matrix I m (In ) by subjecting it to the same elementary row (column) transformation.
Theorem 2 : Let C  AB be a product of two matrices. Any elementary row (column) transformation of
AB can be obtained by subjecting the pre-factor A (post factor B) to the same elementary row (column)
transformation.
Method of finding the inverse of a matrix by elementary transformations : Let A be a non singular
matrix of order n. Then A can be reduced to the identity matrix I n by a finite sequence of elementary
transformation only. As we have discussed every elementary row transformation of a matrix is equivalent to
pre-multiplication by the corresponding elementary matrix. Therefore there exist elementary matrices
E1 , E 2 ..... E k such that
(E k E k 1... E 2 E1 ) A  In
 (Ek Ek 1 ... E 2 E1 ) AA 1  In A 1 (post multiplying by A 1 )
 (Ek Ek 1 ... E 2 E1 ) In  A 1 ( In A 1  A 1 and AA 1  In )  A 1  (Ek Ek 1 ... E 2 E1 ) In
Algorithm for finding the inverse of a non singular matrix by elementary row transformations
Let A be non- singular matrix of order n
Step I : Write A  I n A
Step II : Perform a sequence of elementary row operations successively on A on the LHS and the pre
factor I n on the RHS till we obtain the result I n  BA
Step III : Write A 1  B
Note :  The following steps will be helpful to find the inverse of a square matrix of order 3 by using
elementary row transformations.
Step I : Introduce unity at the intersection of first row and first column either by interchanging two rows or
by adding a constant multiple of elements of some other row to first row.
Step II : After introducing unity at (1,1) place introduce zeros at all other places in first column.
Step III Introduce unity at the intersection of 2 nd row and 2nd column with the help of 2nd and 3rd row.
Step IV : Introduce zeros at all other places in the second column except at the intersection of 2 nd row
and 2nd column.
Step V : Introduce unity at the intersection of 3rd row and third column.
Step VI : Finally introduce zeros at all other places in the third column except at the intersection of third
row and third column.
3  1  2 
 
Example: 37 Using elementary row transformation find the inverse of the matrix A   2 0  1
3  5 0 
 364 Matrices
 5 / 8 5/4 1/8 5 5 1 5 5 1
  1  1 
(a)   3 / 8 3/4  1 / 8  (b)  3 3 1 (c) 3 6 1  (d) None of these
8 8
 5 / 4 3/2 1 / 4  0 3 1 10  12 2 

 3  1  2  1 0 0
   
Solution: (a) We have A=IA   2 0  1   0 1 0 A
3  5 0  0 0 1 

1  1  1  1 1 0
   
Applying (R1  R1  R2 )  2 0  1   0 1 0 A
3  5 0  0 0 1 

1  1  1  1 1 0
   
Applying R2  R2  2R1 and R3  R3  3R1 , 0 2  1    2 3 0 A
0  2 3   3 3 1 

1 1 1   1 1 0
   
Applying R 2  R 2 / 2, 0 1 1 / 2    1 3/2 0 A
0 2 3   3 3 1 

1 0  1 / 2  0 1/2 0
   
Applying R1  R1  R2 and R3  R3  2R2 , 0 1 1 / 2    1 3/2 0 A
0 0 4   5 6 1 

1 0  1 / 2  0 1/2 0 
   
Applying R3  R3 / 4 , 0 1 1 / 2    1 3/2 0 A
0 0 1  5 / 4 6/4 1 / 4 

1 0 0   5 / 8 5/4 1/8
1 1    
Applying R1  R1  R 3 and R 2  R 2  R 3 , 0 1 0    3 / 8 3/4  1 / 8 A
2 2
0 0 1  5 / 4 3/2 1 / 4 

 5 / 8 5/4 1/8
1  
A   3 / 8 3/4  1 / 8
 5 / 4 3/2 1 / 4 

8.2.17 Rank of Matrix .

Definition : Let A be a m×n matrix. If we retain any r rows and r columns of A we shall have a square
sub-matrix of order r. The determinant of the square sub-matrix of order r is called a minor of A order r.
1 3 4 5
Consider any matrix A which is of the order of 3×4 say, A  1 2 6 7 . It is 3×4 matrix so we can have
1 5 0 1
minors of order 3, 2 or 1. Taking any three rows and three columns minor of order three. Hence minor of order
1 3 4
3 1 2 6 0
1 5 0
Making two zeros and expanding above minor is zero. Similarly we can consider any other minor of order
3 and it can be shown to be zero. Minor of order 2 is obtained by taking any two rows and any two columns.
1 3
Minor of order 2   2  3  1  0 . Minor of order 1 is every element of the matrix.
1 2
Rank of a matrix: The rank of a given matrix A is said to be r if
(1) Every minor of A of order r+1 is zero
 Matrices 365

(2) There is at least one minor of A of order r which does not vanish
Note :  If a minor of A is zero the corresponding submatrix is singular and if a minor of A is not zero
then corresponding submatrix is non-singular.
Here we can also say that the rank of a matrix A is said to be r if
(i) Every square submatrix of order r+1 is singular.
(ii) There is at least one square submatrix of order r which is non-singular.
The rank r of matrix A is written as  ( A)  r
Working rule : Calculate the minors of highest possible order of a given matrix A. If it is not zero, then
the order of the minor is the rank. If it is zero and all other minors of the same order be also zero, then
calculate minor of next lower order and if at least one of them is not zero then this next lower order will be the
rank. If, however, all the minors of next lower orders are zero, then calculate minors of still next lower order
and so on.
Note :  The rank of the null matrix is not defined and the rank of every non-null matrix is greater than or
equal to 1.
 The rank of a singular square matrix of order n cannot be n.
8.2.18 Echelon form of a Matrix.
A matrix A is said to be in Echelon form if either A is the null matrix or A satisfies the following conditions:
(1) Every non- zero row in A precedes every zero row.
(2) The number of zeros before the first non-zero element in a row is less than the number of such zeros
in the next row.
It can be easily proved that the rank of a matrix in Echelon form is equal to the number of non-zero row of
0 3 2 1 
the matrix. Example : The rank of the matrix A  0 0 2 5  is 2 because it is in Echelon form and it has
0 0 0 0 
0 2 5

two non-zero rows. The matrix A  0 0 1  is not in Echelon form, because the number of zeros in second
0 0  4 
row is not less than the number of zeros in the third row. To reduce A in the echelon form, we apply some
0 2 5 
elementary row transformations on it. Applying R 3  R 3  4 R 2 , we obtain A ~ 0 0 1  , which is in Echelon
0 0 0 
form and contains 2 non zero rows. Hence, r( A)  2
Rank of a matrix in Echelon form : The rank of a matrix in Echelon form is equal to. the number of non-
zero rows in that matrix.
Algorithm for finding the rank of a matrix : Let A  [a ij ] be an m×n matrix.
Step I : Using elementary row transformations make a11  1
Step II : Make a 21 , a 31 ,...., a m 1 all zeros by using elementary transformations,
R 2  R 2  a 21 R1 , R 3  R 3  a 31 R1 ,..... R m  R m  am 1 R1
Step III : Make a 22  1 by using elementary row transformations.
 366 Matrices
Step IV : Make a 32 , a 42 ,...., a m 2 all zeros by using R 3  R 3  a 32 R 2 , R 4  R 4  a 42 R 2 ,... R m  R m  am 2 R 2
The process used in steps III and IV is repeated upto (m  1)th row. Finally we obtain a matrix in Echelon
form, which is equivalent to the matrix A. The rank of A will be equal to the number of non-zeros rows in it.
2 3 1 4 
 
Example: 38 The rank of the matrix A  0 1 2  1 is [Kurukshetra CEE 2002 ]
0  2  4 2 
(a) 2 (b) 3 (c) 1 (d) Indeterminate
2 3 1 4 2 3 1
Solution: (a) We have A  0 1 
2  1 , Considering 3×3 minor 
 0 1 2
 its determinant is 0.

0  2  4 2  0  2  4 
34 33

2 3 4 2 1 4  3 1 4
     
Similarly considering , 0 1 1  ,  0 2  1 and  1 2  1 , their determinant is 0 each rank can not
0  2 2  0  4 2 
 2  4 2 
be 3
2 3
Then again considering a 2×2 minor,   , which is non zero. Thus, rank =2
0  2 
 1 2 5 
 
Example: 39 The rank of the matrix  2 4 a  4 is [Roorkee 1988]
 1 2 a  1 
(a) 1 if a = 6 (b) 2 if a =1 (c) 3 if a = 2 (d) 1 if a = – 6
 1 2 5  0 0 a6 0 0 0
 
Solution: (b,d) Let A   2 4 a  4  0 0 a6  0 0 a6
 1 2 a  1  1 2 a 1 1 2 a 1
0 0 0
When a  6 , A 0 0 0 ,  r( A)  1
1 2 5
0 0 0 0 0 0
When a  1 , A  0 0 7 r ( A)  2 , When a  6 , A  0 0  12 , r( A)  2
1 2 2 1 2 7
0 0 0
When a  2 , A  0 0 8 ,  r( A)  2
1 2 3
x  a b c 
 
Example: 40 The value of x so that the matrix  a x b c  has rank 3 is
 a b x  c 
(a) x  0 (b) x  a  b  c
(c) x  0 and x  (a  b  c) (d) x  0, x  a  b  c
x a b c
Solution: (c) Since rank is 3, | A | 33  0 , a x b c 0
a b x c 33

x ab c b c
x ab c x b c 0 , Applying (C1  C1  C2  C3 )
x ab c b x c
 Matrices 367

1 b c 1 b c
(x  a  b  c ) 1 x  b c  0  x ab c  0, 1 x b c 0
1 b x c 1 b x c
0 x c
x  (a  b  c) , 0 x x 0  x 0
1 b x c

8.2.19 System of Simultaneous Linear Equations .

Consider the following system of m linear equations in n unknowns
a11 x 1  a12 x 2  ...  a1n x n  b1
a 21 x 1  a 22 x 2  ...  a 2 n x n  b 2
... ... ... ... ...
... ... ... ... ...
a m 1 x 1  a m 2 x 2  ...  a mn x n  b m
 a11 a12 ... a1n   x 1   b1 
    
The system of equations can be written in matrix form as  a 21 a 22 ... a 2 n   x 2   b 2  or AX  B,
        
a m 1 am 2 ... a mn   x n  b n 
 a11 a12 ... a1n  x1   b1 
     
Where A   a21 a22 ... a2 n  , X   x 2  and B  b2 
         
am 1 am 2 ... amn  m n  x n  n1 b m  m 1
The m×n matrix A is called the coefficient matrix of the system of linear equations.
(1) Solution : A set of values of the variables x 1 , x 2 ...... x n which simultaneously satisfy all the equations
is called a solution of the system of equations. Example : x  2, y  3 is a solution of the system of linear
equations 3 x  y  3, 2 x  y  1 , because 3(2) +(–3)=3 and 2(2)+(–3)=1
(2) Consistent system : If the system of equations has one or more solutions, then it is said to be a
consistent system of equations, otherwise it is an inconsistent system of equations. Example : the system of
linear equation 2 x  3 y  5, 4 x  6 y  10 is consistent, because x=1, y =1 and x  2, y  1 / 3 are solutions of
it.
However, the system of linear equations 2 x  3 y  5, 4 x  6 y  10 is inconsistent, because there is no
set of values of x, y which satisfy the two equations simultaneously.
(3) Homogeneous and non-homogeneous system of linear equations: A system of equations AX=B
is called a homogeneous system if B  0 . Otherwise, it is called a non-homogeneous system of equations.
Example : The system of equations, 2 x  3 y  0, 3 x  y  5 is a homogeneous system of linear equations
whereas the system of equations given by 2 x  3 y  1, 3 x  y  5 is a non homogeneous system of linear
equations.
8.2.20 Solution of a Non Homogeneous System of Linear Equations.
There are three methods of solving a non homogeneous system of simultaneous linear equations.
(1) Determinant Method (Cramer's Rule) (2) Matrix method (3) Rank method
We have already discussed the determinant method (Cramer's rule) in chapter determinants.
 368 Matrices
(1) Matrix method : Let AX  B be a system of n linear equations with n unknowns. If A is non-singular,
then A 1 exists.  AX  B  A 1 ( AX )  A 1 B , [pre-multiplying by A 1 ]
 ( A 1 A)X  A 1 B , [by associativity]
 In X  A 1 B  X  A 1 B . Thus, the system of equations AX  B has a solution given by X  A 1 B .
Now, let X 1 and X 2 be two solutions of AX  B . then, AX 1  B and AX 2  B
1 1 1 1
 AX 1  AX 2  A ( AX 1 )  A ( AX 2 )  ( A A)X 1  ( A A)X 2  In X 1  In X 2  X 1  X 2 .
Hence, the given system has a unique solution.
Thus, if A is a non-singular matrix, then the system of equations given by AX  B has a unique solution
given by X  A 1 B .
If A is a singular matrix, then the system of equations given by AX=B may be consistent with infinitely
many solutions or it may be inconsistent also.
Criterion of consistency : Let AX  B be a system of n-linear equations in n unknowns.
(i) If | A |  0, then the system is consistent and has a unique solution given by X  A 1 B
(ii) If | A |  0 and (adj A) B  0 , then the system is consistent and has infinitely many solutions.
(iii) If | A |  0 and (adj A) B  0 , then the system is inconsistent
Algorithm for solving a non-homogeneous system of linear equations : We shall give the algorithm
for three equations in three unknowns. But it can be generalized to any number of equations.
Let AX  B be a non-homogenous system of 3 linear equations in 3 unknowns. To solve this system of
equations we proceed as follows
Step I : Write the given system of equations in matrix form, AX  B and obtain A, B.
Step II : Find | A |
Step III : If | A |  0 , then write "the system is consistent with unique solution". obtain the unique solution
1
by the following procedure. Find A 1 by using A 1  adj A obtain the unique solution given by X  A 1 B
| A|
Step IV : If | A |  0, then write "the system is either consistent with infinitely many solutions or it is
inconsistent. To distinguish these two, proceed as follows: Find (adj A) B.
If (adj A)B  0 , then write "the system is inconsistent”.
If (adj A)B  0 , then the system is consistent with infinitely many solution. To find these solutions proceed
as follows. Put z  k (any real number) and take any two equations out of three equations. Solve these
equations for x and y. Let the values of x and y be  and  respectively. Then x   , y   z  k is the
required solution, where any two of , , k are functions of the third.
(2) Rank method : Consider a system of m simultaneous linear equations in n unknowns x 1 , x 2 ...., x n ,
given by a11 x 1  a12 x 2  a13 x 3  .......  a1n x  b 1
a 21 x 1  a 22 x 2  a 23 x 3  .......  a 2n x n  b 2
    

a m 1 x 1  a m 2 x 2  a m 3 x 3  ....  a mn x n  b m
This system of equations can be written in matrix form as
 Matrices 369

 a11 a12 a13 .... a1 n   x 1   b1 

     
 a 21 a 22 a 23 .... a 2n  x 2   b2 
           
a m 1 am 2 am 3 .... a mn   x n  b m 
 a11 a12 a13 .... a1n  x1   b1 
     
or AX  B ,where  a 21 a 22 a 23 .... a 2 n  , X   x 2  and B   b 2 
           
a m 1 am 2 am 3 .... amn  m n  x n  n1 am  m 1
The matrix A is called the coefficient matrix and the matrix
 a11 a12 a13 .... a1n b1 
 
[ A : B]   a 21 a 22 a 23 .... a 2 n b 2  is called the augmented matrix of the given system of
     
a m 1 a m 2 a m 3 .... a mn b m 
equations. This matrix is obtained by adding (n  1) column to A. The elements of this column are
b1 , b2 ...., bm
For example, the augmented matrix of the system of equation
2x  y  3z  1
x  y  2z  5
x  y  z  1 is
2  1 3 1
1 1 2 5 

1 1 1  1
A non-homogeneous system of linear equations may have a unique solution, or many solutions or no
solution at all. If it has a solution (whether unique or not) the system is said to be consistent. Otherwise it is
called an inconsistent system. The following theorems tell us about the condition for consistency of a system of
linear equations
Theorem 1 : The system of linear equations AX  B is consistent iff the rank of the augmented matrix
[ A : B] is equal to the rank of the coefficient matrix A.
Theorem 2 : Let AX  B be a system of m simultaneous linear equations in n unknowns.
Case I : If m  n , then
(i) if r ( A)  r ( A : B)  n, then system of linear equations has a unique solution.
(ii) if r ( A)  r ( A : B)  r  n, then system of linear equations is consistent and has infinite number of
solutions. In fact, in this case (n  r) variables can be assigned arbitrary values.
(iii) if r ( A)  r ( A : B), then the system of linear equations is inconsistent i.e. it has no solution.
Case II : If m  n and r ( A)  r( A : B)  r, then r  m  n and so from (ii) in case I, there are infinite number
of solutions.
Thus, when the number of equations is less than the number of unknowns and the system is consistent,
then the system of equations will always have an infinite number of solutions.
Algorithm for solving a non-homogeneous system AX=B of linear equations by rank method
Step I: Obtain A, B.
Step II : Write the Augmented matrix [A : B].
Step III : Reduce the augmented matrix to Echelon form by applying a sequence of elementary row-
operations.
Step IV : Determine the number of non-zero rows in A and [A : B] to determine the ranks of A and [A :B]
respectively.
 370 Matrices
Step V: If r ( A)  r( A : B) then write "the system is inconsistent" STOP else write "the system is
consistent", go to Step VI
Step VI : If r ( A)  r ( A : B) = number of unknowns, then the system has a unique solution which can be
obtained by back substitution.
If r ( A)  r ( A : B) < number of unknowns, then the system has an infinite number of solutions which can
also be obtained by back substitution.
 x  y 2 x  z  4 7 
Example: 41 If    then values of x, y, z, w are
 x  y 2 z  w  0 10 

(a) 2, 2, 3, 4 (b) 2, 3, 1,2 (c) 3, 3, 0, 1 (d) None of these

 x  y 2 x  z  4 7 
Solution: (a) We have   
 x  y 2 z  w  0 10 

x  y  4 , 2 x  z  7 , x  y  0 and 2 z    10  x  2 and y  2, z  3, w  4

Example: 42 The system of linear equation x  y  z  2 , 2 x  y  z  3, 3 x  2 y  kz  4 has unique solution if [EAMCET 1994]

(a) K  0 (b) 1  K  1 (c) 2  K  2 (d) K  0

1 1 1
Solution: (a) The given system of equation has a unique solution if 2 1  1  0  K  0
3 2 k

8.2.21 Cayley-Hamilton Theorem .

Every matrix satisfies its characteristic equation e.g. let A be a square matrix then | A  xI |  0 is the
characteristics equation of A. If x 3  4 x 2  5 x  7  0 is the characteristic equation for A, then
A3  4 A 2  5 A  7 I  0
Roots of characteristic equation for A are called Eigen values of A or characteristic roots of A or latent roots
of A.

If  is characteristic root of A, then 1 is characteristic root of A 1 .

8.2.22 Geometrical Transformations .
(1) Reflexion in the x-axis: If P ' (x ' , y ' ) is the reflexion of the point P( x , y ) on the x-axis, then the matrix
1 0 
0  1 describes the reflexion of a point P( x , y ) in the x-axis.
 
 1 0 
(2) Reflexion in the y-axis : Here the matrix is  
 0 1
 1 0 
(3) Reflexion through the origin : Here the matrix is  
 0  1
0 1 
(4) Reflexion in the line y = x : Here the matrix is  
1 0 
 Matrices 371

 0  1
(5) Reflexion in the line y = – x : Here the matrix is  
 1 0 
cos 2 sin 2 
(6) Reflexion in y = x tan  : Here matrix is 
 sin 2  cos 2 

cos   sin  
(7) Rotation through an angle  : Here matrix is 
 sin  cos  

8.2.23 Matrices of Rotation of Axes .

We know that if x and y axis are rotated through an angle  about the origin the new coordinates are
given by x  X cos   Y sin  and y  X sin   Y cos 

 x  cos   sin    X  cos   sin  

   is the matrix of rotation through an angle  .
 y   sin  cos    Y   sin 
 cos  

1 1 3 
 
Example: 43 Characteristic equation of the matrix A   1 3  3
 2  4  4 

(a) A 3  20 A  8 I (b) A 3  20 A  8 I (c) A 3  80 A  20 I (d) None of these

Solution: (a) The characteristic equation is | A  I |  0 .

1 1 3
So, 1 3   3  0 i.e. 3  20   8  0
2 4 4 

By cayley-Hamilton theorem , A 3  20 A  8 I  0
Example: 44 The transformation due to the reflection of (x , y ) through the origin is described by the matrix

0 0  1 0   0 1 1 0 
(a)   (b)   (c)   (d)  
0 1   0  1  1 0  0 1 
Solution: (b) If x , y  is the new position

x   (1)x  0 .y, y   0 . x  (1)yd

 x  1 0   x 
      
 y   0  1  y 

1 0 
 Transformation matrix is  
 0  1

Example: 45 The rotation through 180 0 is identical to

(a) The reflection in x -axis (b) The reflection in y-axis (c) A point reflection (d) Identity transformation

Solution: (c) Rotation through 180 0 gives x    x

y    y . Hence this a point reflection.
372 Matrices
***