MATRICES

1.1. MATRIX
A matrix over a field F is a collection of mn elements of a field F arranged in the form of
a rectangular array having m rows and n columns and called an m  n matrix over the
field F . The matrix is called real matrix or complex matrix according as the field is of
real numbers or complex numbers.

Throughout the present treatment, the elements of a matrix shall be assumed to be

complex numbers unless stated otherwise.

An m  n matrix is usually written as

 a11 a12 ... a1n 

a ... a 2n 
 21 a 22
A  a 31 a 32 ... a 3n 
 
    
am1 am 2 ... amn 


 In a compact form the above matrix is represented by A  aij m n .

 In the element a ij the first suffix i will always denote the row and the second suffix j
, the column in which the element occurs.

1.1.1. OPERATIONS ON MATRICES

1. Addition: Let A and B be two matrices of the same type m  n . Then their sum
A  B is defined to be the matrix of the type m  n obtained by adding the
corresponding elements of A and B .

Thus if A  aij  and B  bij m n , then A  B  aij  bij m n

m n

 If two matrices A and B are of the same size, they are said to be conformable
for addition, if the matrices A and B are not of the same size, we cannot find
their sum.

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 1

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

Properties of Matrix Addition

(i) Matrix addition is Commutative: If A and B be two m  n matrices, then

A B B  A.

(ii) Matrix addition is associative: If A, B , C be three matrices each of the type m  n ,

then  A  B   C  A  B  C  .

(iii) Existence of additive identity: If O be the m  n matrix each of whose entries is

zero, then A  O  A  O  A for every m  n matrix A .

(iv) Existence of additive inverse: Let A  aij m n ,then the negative of the matrix A is
defined as the matrix  aij m n and is denoted by A .

(v) Cancellation laws hold good in the case of addition of matrices: If A, B , C are
three m  n matrices, then

A  B  A C  B C (Left cancellation law)

B  A C  A B C (Right cancellation law)

2. Scalar Multiplication: Let A be any m  n matrix and k any element from field (called
scalar). The m  n matrix obtained by multiplying every element of the matrix A by k is
called the scalar multiple of A by k and is denoted by kA or Ak . Symbolically, if
A  aij  , then kA  Ak  kaij  .
m n m n

Properties of scalar multiplication:

(i) If A and B are two matrices each of the type m  n , then k  A  B   kA  kB .

(ii) If A be any m  n matrix and k be any scalar, then  k  A   kA   k   A  .

3. Matrix multiplication: Let A  aij m n and B  bij n  p be two matrices then their
n
product AB is a m  p matrix C  c ij m  p such that c ij   aik bkj  ai1b1 j  a12b2 j  ...  ain bnj .
k 1

 The i , k  element cik of the matrix AB is obtained by multiplying the

th

corresponding elements of the i th row of A and the k th column of B and then

adding the products.

 The product AB of two matrices A and B exists if and only if the number of
columns in A is equal to the number of rows in B . Two such matrices are said
to be conformable for multiplication.

 If the product AB exists, then it is not necessary that the product BA will also
exist.

2 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 Properties of Matrix Multiplication:

(i) Matrix multiplication is associative i.e. if A, B , C are m  n , n  p, p  q matrices

respectively then A  BC    AB  C .

(ii) Multiplication of matrices is distributive with respect to addition of matrices

i.e. A  B  C   AB  AC where A, B , C are any three, m  n , n  p, n  p matrices
respectively

(iii) The multiplication of matrices is not always commutative:

(a) Wherever AB exist, it is not always necessary that BA should also exist.

(b) Wherever AB and BA both exist, it is always not necessary that they should be
matrices of the same size.

(c) Whenever AB and BA both exist and are matrices of the same size, it is not
necessary that AB  BA .

 The matrices A and B are said to commute if AB  BA . If AB  BA , the matrices A
and B are said to anti-commute.

(iv) If A be any m  n matrix and On  p be an n  p matrix with all entries zero, then
AOn  p  O m  p where Om  p is an m  p matrix with all entries zero,similarly if Om n be a
m  n matrix with all entries zero and A be any n  p matrix, then Om n A  Om  p .

(v) The equation AB  O does not necessarily imply that at least one of the matrices A
and B must be a zero matrix.

(vi) In the case of matrix multiplication if AB  O , then it does not necessarily imply
that BA  O .

(vii) If A be a m  n matrix and I n denotes the n-rowed unit matrix (defined on next
page), it can be easily seen that AI n  A  I m A .

(viii) A useful ways of representing matrix products:

Let A  aij mn , B  bij n  p and R1, R 2 ,..., Rm be the rows of A and C1 , C 2 ,..., C p be the columns
of B then

R1   R1C1 R1C 2  R1C p 

R  R C R C  R2C p 
 2  2 1 2 2

AB   C1 C 2  C P  
       
   
       
R  RmC1 RmC 2  RmC p 
 m 

b1 j 
 
b2 j
Where RiC j  ai1 ai 2  ain   

 
bnj 

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 3

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

1.1.2. TYPES OF MATRICES

1. Square Matrix: A m  n matrix for which m  n (i.e., the number of rows is equal to
the number of columns) is called a square matrix of order m . The elements a ij of a
square matrix A  aij n n , for which i  j i.e., the elements a11 , a 22 , a 33 ,..., ann are called
the diagonal elements and the line along which they lie is called the principal
diagonal of the matrix.

2. Diagonal Matrix: A square matrix A  aij n n whose elements above and below the
principal diagonal are all zero i.e. aij  0 for i  j , is called a diagonal matrix.

 An n-rowed diagonal matrix whose diagonal elements in order and d1 , d2 , d3 ,..., dn will
often be denoted by the symbol Diag d1 , d 2 ,..., dn  or d d1 , d2 ,..., dn  .

3. Scalar Matrix: A diagonal matrix whose diagonal element are all equal is called a
scalar matrix.

 The pre-multiplication or the post multiplication of A by a scalar matrix has the

same effect as the multiplication of A by the scalar k . This is perhaps the
motivation behind the name scalar matrix.

4. Unit Matrix or Identity Matrix: A square matrix each of whose diagonal elements
is 1 and each of whose non– diagonal elements is equal to zero is called a unit
matrix or an identity matrix and is denoted by I , I n will denote a unit matrix of order
n. Thus a square matrix A  aij  is a unit matrix if aij  1 when i  j and aij  0 when
i  j.

5. Null Matrix or Zero Matrix: The m  n matrix whose elements are all 0 is called the
null matrix (or zero matrix) of the type m  n . It is usually denoted by O or more

6. Row Matrices or Column matrices: Any 1  n matrix which has only one row and n
columns is called a row matrix or a row vector. Similarly any m  1 matrix which has
m rows and only one column is a column matrix or a column vector.

7. Equal matrices: Two matrices A  aij  and B  bij  are said to be equal if they are of
the same size and the elements in the corresponding places of the two matrices are
the same i.e., a ij  bij for each pair of subscripts i and j .If two matrices A and B are
equal, we write A  B .

8. Triangular Matrix:

(i) Upper Triangular Matrix : A square matrix A  aij  is called an upper triangular
matrix if aij  0 wherever i  j .

(ii) Lower Triangular Matrix: A square matrix A  aij  is called a lower triangular
matrix if aij  0 whenever i  j .

 Thus a diagonal matrix is both upper and lower triangular.

 A triangular matrix is called strictly triangular if aii  0 for all 1  i  n .

4 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 Problem: The product of two triangular matrices is itself triangular both should be
upper triangle or lower triangle.

Solution: Let A  aij n n and B  b jk n n be two triangular matrices each of order n . Then
aij  0 when i  j and b jk  0 when j  k .

n
Let AB  c ik n n , where c ik   aij b jk .
j 1

Suppose that i  k .

If j  i , then aij  0 and therefore c ij  0 .

If j  i , then j  k because i  k . In this case bij  0 and therefore c ik  0

Thus c ik  0 whenever i  k .

Hence the matrix AB is also a triangular matrix.

Problem: The pre-multiplication (or post multiplication) of a square matrix A by a

diagonal matrix D multiplies each row (or column) or A by the corresponding diagonal
element of D and hence the only matrices commutative with a diagonal matrix with
distinct diagonal elements are diagonal matrices.

Solution: Let A  aij n n be a square matrix.

b11 0 0  0 
a11 a12  a1n  0 b
a   0  0 

12
a a
i.e. A   21 22 2n 
and D   0 0 0  0  be a diagonal matrix of order n .
      

a a  a
      
 n1 n2 nn   0 0 0  bnn 

 b11a11 b11a12  b11a1n 

b a b12a 22  b12a 2n 
Then DA   12 11

    
 
bnn an1 bnn an 2  bnn ann 

Thus in the product DA the first row of A has been multiplied by the corresponding
diagonal element b11 of the first row of D and so on.

a11b11 a12b22  a1n bnn 

a b a 22b22  a12bnn 
Again AD   11 11
    
 
an1b11 an 2b22  ann bnn 

For this it is obvious that in the product AD the first column of A has been multiplied
by the corresponding diagonal element b11 of the first column of D and so on.

Hence the given result is true for post multiplication of A by D .

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 5

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

Second Part: Let A  aij n n commute with the diagonal matrix D of order n having its
diagonal elements all distinct. Let D  d b11 ,..., bnn 

The i , j  element of DA  a ij bii and the i , j  element of AD  aij b jj .

th th

But AD  DA , so a ij bii  a ij b jj or aij bii  b jj   0 .

Given if i  j , then bii  b jj , therefore aij  0 if i  j .

Thus each non-diagonal element of A is zero and A is a diagonal matrix.

Problem: If a diagonal matrix is commutative with every matrix of the same order then
it is necessarily a scalar matrix.

Solution: Let D  d b11 , b22 ,..., bnn  be a diagonal matrix of order n . Let A  aij n n be any
square matrix of order n .

The i , j  element of DA  a ij bii and the i , j  element of AD  aij b jj

th th

Now given that AD  DA , therefore a ij bii  a ij b jj or aij bii  b jj   0

Since A is any square matrix of order n , therefore we can take aij  0 . Therefore we
must have bii  b jj  0 or bii  b jj for each i and j .

Thus the diagonal elements of D are all equal. Therefore D is a scalar matrix.

1.1.3. SOME OPERATIONS ON MATRIX

1. Trace of a Matrix: Let A be a square matrix of order n . The sum of the elements of
A lying along the principal diagonal is called the trace of A . We shall write the trace
n
of A as tr A . Thus if A  aij n n then tr A   a ii  a11  a 22  ...  ann .
i 1

Properties of Trace

Let A and B be two square matrices of order n and k be a scalar. Then

(i) tr   A    trA

(ii) tr  A  B   tr A  trB

(iii) tr  AB   tr  BA 

(iv) tr  ABC   tr CAB   tr  BCA  but tr  ABC   tr  ACB  .

a
2
(v) For a matrix A  [aij ]nn , tr ( At A)  tr ( AAt )  ij .
i, j

6 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 (vi) For a real matrix A  [aij ]nn , tr ( A A)  0 and tr ( A A)  0 iff A  0 .
t t

2. Transpose of a Matrix: Let A  aij m n then the n  m matrix obtained from A by
changing its rows into columns and its column into rows is called the transpose of
A and is denoted by the symbol At or AT .

 The operation of interchanging rows with columns is called transposition.

Symbolically if A  aij m n , then At  b ji n m , where b ji  a ij , i.e. the  j , i 

th

element of At is the i , j  element of A .
th

Properties of Transpose

If A t and B t be the transpose of A and B respectively. Then

(i) A t t
A

A  B 
t
(ii)  At  B t , A and B being of the same size.

(iii) kA   kA t , k being scalar.

t

(iv)  AB   B t At , A and B being conformable for multiplication.

t

3. Conjugate of a matrix: The matrix obtained from any given matrix A on replacing
its elements by the corresponding conjugate complex number is called the
conjugate of A and is denoted by A .

 Thus if A  aij m n , then A  aij m n where a ij denotes the conjugate complex of a ij .

 If A be a matrix over the field the real number, then obviously A coincides with A .

Properties of Conjugate

If A and B be the conjugates of A and B respectively, then

(i)  A   A

(ii) A  B  A  B

(iii) kA   k A , k being any scalar

(iv) AB  AB , A and B being conformable for multiplication.

4. Transposed conjugate of a matrix: The transpose of the conjugate of a matrix A is

called transposed conjugate of A and is denoted by A or by A * and we call it
tranjugate of A .

Properties of Conjugate: If A and B  be the transposed conjugates of A and B

respectively, then

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 7

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

(i)  A '   A  '  A

(ii)  A   A


(iii)  A  B   A  B  , A and B being of the same size



(iv) kA   kA ,k being any complex number



(v)  AB   B  A , A and B being conformable to multiplication.



1.1.4. SOME MORE SPECIAL MATRICES

1. Symmetric matrices: A square matrix A  aij  is said to be symmetric if its i , j 

th

element is the same as its  j , i  element i.e. if aij  a ji for all i , j .

th

 If A and B are symmetric matrices, then show that AB is symmetric if and only if
A and B commute i.e. AB  BA .
t t
 If A be any matrix, then prove that AA and A A are both symmetric matrices.

If A be any square matrix, then show that A  A is symmetric and A  A is skew-

t t

symmetric.

2. Skew symmetric matrix: A square matrix A  aij  is said to be skew symmetric if

the i , j  element of A is the negative of the  j , i  element of A i.e. if aij  a ji for all
th th

i, j .

 If A is a skew –symmetric matrix, then aij  a ji for all i , j [by definition]

 aii  aii for all values of i

 2aii  0 or aii  0 .

Thus the diagonal elements of a skew-symmetric matrix are all zero.

 If A is a symmetric (skew-symmetric) matrix, then show that kA is also symmetric

(skew-symmetric).

 If A, B are symmetric (skew-symmetric) , then so is also A  B .

 Positive odd integral power of a skew-symmetric matrix is skew-symmetric while

positive even integral powers are symmetric.

 A2 is symmetric if either A is symmetric or A is skew-symmetric.

 If A and B are skew-symmetric matrices of order n , then show that AB is

symmetric if and only if A and B commute.

8 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
  If A and B are symmetric matrices of order n , then show that AB  BA is symmetric
and AB  BA is skew-symmetric.

 Every square matrix is uniquely expressible as the sum of a symmetric matrix and
1
a skew-symmetric matrix using A 
2
 A  At   12  A  At  .

3. Hermitian Matrix: A square matrix A  aij  is said to be Hermitian if the i , j 

th

element of A is equal to the conjugate of the  j , i  element of A i.e. if aij  a ji for all
th

i,j.

 Thus every diagonal element of a Hermitian matrix must be real.

 A Hermitian matrix over the field of real numbers is nothing but a real symmetric
matrix.

1
 Every Hermitian matrix A can be written as A  B  iC where B  A  A  is real
2
1
and symmetric and C  A  A  is real and skew-symmetric using
2i
1
A
2
 A  A   21i  A  A  .
 If A is Hermitian, such that A 2  O , then A  O , where O is the zero matrix.

4. Skew-Hermitian Matrix: A square matrix A  aij  is said to be skew-Hermitian if

the i , j  element of A is equal to the negative of the conjugate of the  j , i  element
th th

of A i.e., if aij  a ji for all i , j .

 The diagonal elements a skew – Hermitian matrix must be pure imaginary numbers
or zero.

 If A is a Hermitian matrix, then iA is skew-Hermitian and if A is a Skew –Hermitian

matrix, then iA is Hermitian.

 If A, B are Hermitian or Skew- Hermitian, then so is also A  B .

 If A, B are Hermitian; then AB  BA is Hermitian and AB  BA is skew-Hermitian.

 
 If A be any square matrix, then A  A , AA , A A are all Hermitian and A  A is Skew
–Hermitian.

 Every square matrix is uniquely expressible as the sum of a Hermitian matrix and a
1 1
skew-Hermitian matrix using A 
2
 A  A    A  A   .
2

5. Idempotent Matrix: A matrix A such that A2  A is called idempotent.

 d d1 , d2 ,..., dn  , d1, d2 ,..., dn  0,1 are the only Idempotent diagonal matrix of order n.

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 9

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

1 a 
 The matrix   is Idempotent for every a  .Thus there are infinite Idempotent
0 0 
matrices.

 If AB  A and BA  B , then A and B are idempotent.

 If B is an idempotent matrix, then A  I  B is also idempotent and AB  BA  O .

6. Involutory Matrix: A matrix A such that A2  I is called involutory.

 A isinvolutory, if and only if  I  A   I  A   O .

1 a 
 The matrix   is Involutory for every a  .Thus there are infinite Involutory
0 1
matrices.

7. Nilpotent Matrix: A square matrix A is called a nilpotent matrix if there exists a

positive integer m such that A m  O . If m is the least positive integer such that
A m  O , then m is called the index of the nilpotent matrix A .

 ab b2 
 The matrix A    ,where a , b   ,is the only nilpotent matrix of order 2.
 a ab 
2

8. Orthogonal Matrix: A square matrix A is a said to orthogonal if At A  I .

 If A, B be n-rowed orthogonal matrices, then AB and BA are also orthogonal

matrices.

 Every 2-rowed real orthogonal matrix is of any one of the forms

cos   sin   cos  sin  

 sin   or  .
 cos    sin   cos  

t 1
 If A is an orthogonal matrix, then A and A are also orthogonal.

9. Unitary Matrix: A square matrix A is said to be unitary if A A  I .

 If A, B be n-rowed unitary matrices, then AB and BA are also unitary matrices.

 A unitary matrix over the field of real number is orthogonal i.e. a real unitary
matrix is an orthogonal matrix.

10. Normal Matrix: A square matrix A is called normal if AA  A A .

 A normal triangular matrix is diagonal

 If A and B are normal with AB = BA, then both AB and A + B are also normal.

 A Hermitian, Skew-Hermitian, Unitary and Diagonal Matrices are always Normal.

10 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
1.2. DETERMINANTS
Let F be a field and A be a matrix over field F then the determinant of A is an element
of F defined by
n
det A 

 sgn   a 
Sn i 1
i (i )
.

Here, Sn denotes that set (in fact, a group) of all bijections from the set of integers
 ( j )   (i )
1,...,n  to itself and sgn  is the signature of  , defined by sgn    (which is
1i  j n j i
+1 or -1) .

 Determinant of A is also written as | A | .

 Singular and Non-singular matrices: A square matrix A is said to be non-singular

or singular according as A  0 or A  0 .

 If the product of two non-zero square matrices is a zero matrix, then both of them
must be singular matrices.

 If A is idempotent and A  I , then A is singular.

EXAMPLE:

a a 
(i) For A   11 12  , det A  sgn I a11a22  sgn(12)a12a 21  a11a 22  a12a21 .
a 21 a 22 

a11 a12 a13 

(ii) For A  a21 a22 a23  , det A  sgn I a11a 22a 230  sgn 12 3  a12a 23a 31
a31 a32 a33 
 sgn 13 2  a13a 21a 32  sgn 13  a13a 22a 31  sgn 12  a12a 21a 33  sgn  3 2  a11a 23a 32

 a11a22a 33  a12a 23a31  a13a21a 32 a13a 22a 31  a12a 21a 33  a11a 23a 32 .

Properties of Determinant:

(i) If A be a square matrix of order n , then

(a) At  A (b) A  A (c) A  A

(ii) If A and B be two square matrices of the same order, then AB  A B .

(iii) If any two rows (or two columns) of a determinant are interchanged, the value of
the determinant is multiplied by -1.

(iv) If all the elements of one row (or one column) of a determinant are multiplied by
the same number k , the value of the new determinant is k times the value of the
given determinant.

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 11

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

(v) If all the element of a row (or a column) of a determinant are zero, the value of
the determinant is zero.

(vi) If A be an n-rowed square matrix, and k be any scalar, then kA  k n A .

(vii) If two rows (or two columns) of a determinant are identical, the value of the
determinant is zero.

Problem: If A is a skew symmetric matrix of odd order say 2k  1 , then det A  0 .

Solution: As det A  det At

 det A  det(  A) as At   A

 det A  (1)2k 1 det( A) as det(kA )  k n det A

 det A   det A

 det A  0

1.3. INVERSE AND RANK OF A MATRIX

1.3.1 SOME IMPORTANT DEFINITIONS
1. Sub-matrix of a matrix: Suppose A is any matrix of the type m  n . Then a matrix
obtained by deleting some row and column from A is called a sub-matrix of A . In
particular the matrix A itself is a sub-matrix of A because it is obtained from A by
deleting no rows or columns.

2. Minors of a Matrix: We know that every square matrix possesses determinant. If

A be an m  n matrix, then the determinant of every square sub-matrix of A is called
a minor of the matrix A . If we delete m  p rows and n  p columns from A , we
shall get a square sub-matrix of A of order p . The determinant of this square sub-
matrix is called a p -rowed minor of A or minor of order p.

3. Co-factors of a Matrix: Let A be a square matrix of order n, and A(i , j ) be a sub-

th
matrix of A obtained by removing i th row and j column. Then co-factor of the
i j
element aij is (1) det A(i, j ) and we denote it by Aij . Thus a n  n matrix has n 2 co-
factors.

 If A is a square matrix of order n then

n n
(i)  aij Aij  det A (i  1, 2,..., n) (ii)  aij Aij  det A ( j  1, 2,..., n)
j 1 i 1

n n
(iii) a A
i 1
ij kj  0 for i  k (iv) a A
i 1
ij ik  0 for j  k

12 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 4. Adjoint of a square matrix: Let A  aij n n be any n  n matrix. Then the transpose
of the matrix B   Aij n n , where Aij denotes the cofactor of the element a ij of the
matrix A , is called the adjoint of the matrix A and is denoted by the symbol adj A .

 Sometimes the adjoint of a matrix is also called the adjugate of the matrix.

 If A  [aij ] ,and Adj  [ Aij ] then

 n n n

  a1 j Aj1  a1 j Aj 2  a Ajn 
1j
 j 1 j 1 j 1
 | A | 0  0 
 n n n 
  a2 j Aj1 a Aj 2   a2 j A jn   0 | A |  0 
[Link]   j 1
2j
  | A | I n
j 1 j 1     
      
 n n n
  0 0  | A |
 a A   anj A jn 
 nj j1 a nj Aj 2
 j 1 j 1 j 1 

 [Link]  adjA. A | A | I n

 As [Link]  adjA. A | A | I n

 | [Link] ||| A | I n |

 | A || adjA || A |n | I n || A |n

 | adjA || A |n 1

Example: If A is a non-singular matrix, then show that adj adj A   A

n 2
A.

Solution: We have A adj A   A I n . …. (1)

It we take adj A in place of A , then (1) gives adjA adj (adj A )  adjA I n

n 1
adj A  adj adj A   A
n 1
In as adj A  A

Pre –multiplying both sides of this last relation by A , we get

A adj A adj adjA   A. A

n 1
In

  A adjA  adj (adjA )  A  AI n  as matrix multiplication is associative

n 1

  A I n  adj (adjA )  A
n 1
A

n 2
 adj adj A  A A as A is non-singular.

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 13

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

If A is a square matrix of order n , prove that adj adj A   A n 1 .

2


A 0 adj A  0
 If , prove that .

 If A and B are square matrices of the same order, then adj  AB   adj B .adj A .

 If A be a square matrix, then adj At  adj A  .

t

 If A is a symmetric matrix, then prove that adj A is also symmetric.

If A be a square matrix, then show that adjA  adj A  .




 If A is a Hermitian matrix, then prove that adjA is also Hermitian

5. Invertible Matrices or Reciprocal of a matrix: Let A be any n-rowed square

matrix. Then a matrix B , if it exists, such that AB  BA  I n is called inverse of A
and then A is called invertible.

 If B is inverse of A then we denote B as A 1 .

 If A, B be two n-rowed invertible matrices, then AB is also invertible and

 AB 
1
 B 1A 1 . (Reversal Law)

If A be an n  n non-singular matrix, then  At    A 1  and  A 1    A  .

1 t  1


 If the non-singular matrix A is symmetric, then A 1 is also symmetric.

If A is a non-singular matrix, then det  A 1    det A  .

1


1
 Now as [Link]  adjA. A | A | I n , so if | A | 0 then A1  adjA .
| A|

1.3.2. RANK OF A MATRIX

Definition: A number r is said to be the rank of a matrix A if it possesses the following

two properties:

(a) There is at least one non-zero minor A of order r .

(b) All the minors of size r  1 must be zero.

In short the rank of a matrix is the order of any highest order non-vanishing minor
of the matrix.

 We shall denote the rank of a matrix A by the symbol   A  .

14 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
  It is obvious that the rank r of an m  n  matrix can at most be equal to the smaller
of the numbers m and n , but it may be less.

 Thus the rank of every non-singular matrix of order n is n as it has no minor of size
n  1 . The rank of a square matrix A of order n can be less than n if and only if A is
singular i.e., det A  0 .

 The rank of the transpose of a matrix is the same as that of the original matrix.

 The rank of a product of two matrices cannot exceed the rank of either matrix.

  ( A  B )   ( A)   ( B )

  ( A  B )  |  ( A)   ( B ) |

 ( AB)  min   ( A),  ( B)

 .

 Sylvester’s Rank Inequality: If A is a m  n matrix and B is a n  p matrix than

 ( A)   ( B )  n   ( AB ) .

 Rank of AB and rank of BA need not be same.

1 0  0 1 
For example if A    and B    then  ( AB)  1 but  ( BA)  0 .
0 0  0 0 

 If P, A, Q are matrices of size m  m, m  n, n  n respectively and P, Q are invertible

then  ( PA)   ( AQ )   ( PAQ )   ( A) .

 If A is a real matrix then  ( A)   ( At )   ( AAt )   ( At A) .

Problem: If A be an n  n matrix, then rank of adjA is n ,1 or 0 according as the rank

of A is n , n  1 or less than n  1 .

Solution:

(i) Let A be an n  n matrix. Then A  Adj A   A I n .

 A Adj A  A I n  An

Since the matrix A is of rank n , therefore A  0 .

 A Adj A  A gives Adj A  1 .

Thus the matrix adjA is also non-singular. Hence it is of rank n .

DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 15

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
 CSE (Mains) : Mathematics (Optional) - Study Material

(ii) If the rank of A is n  1 , then at least one minor of order n  1 of the matrix A is not
equal to zero. Therefore the matrix adjA . A will be a non-zero matrix and thus the
rank of the matrix adjA will be greater than zero.

Again the rank of the matrix A is n  1 so det A  0 , therefore A adjA  is a zero

matrix and of rank zero.

Now  (A )   (adjA )  n   ( [Link] )  0

 n  1   (adjA )  n

  (adjA )  1

But we have shown that (A)  0

Hence  (A )  1 .

(iii) If the rank of A is less than n  1 , then all minors of order n  1 of the matrix A will
be zero. Therefore the matrix adjA will be a zero matrix and hence  (adjA ) will be
zero.

1.3.3. ELEMENTARY OPERATIONS ON MATRICES

The following operations on row (respectively, columns) of a matrix are called
elementary row (respectively, column) operations:

I. Interchange of two rows (respectively, columns);

II. Multiplication of a row (respectively, column) by a nonzero scalar;

III. Adding to a row (respectively, column) a scalar multiple of another row

(respectively, column)

 Every elementary row (respectively, column) operations on an m  n matrix A can

be performed by pre multiplying (respectively, post multiplying) A by a suitable
m  m elementary (respectively, n  n ) matrix.

 Define the following n  n square matrices:

(i)  r , s   I n  Ers  E sr  Err  E ss , the matrix obtained by interchanging the r  th and

s  th row of the identity matrix, called an elementary permutation matrix.

(ii)  r     I n     1 Err ,   0 , called an elementary dilation.

(iii) rs     I n  Ers , r  s , called an elementary transvection.

Where Ers  [ aij ]n such that aij  0 when i  r , j  s and ars  1

 We call the above matrices as elementary matrices.

16 DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7
  By definition when a single row(column)operation is applied to the identity matrix
then the resultant matrix is called Elementary matrix.

 Elementary matrices are invertible and the inverse of an elementary matrix is an

elementary matrix of the same type and further  r , s    r , s  ,  r      r   1  ,
1 1

 rs      rs    .
1

 A matrix which is product of elementary permutation matrices is called a

permutation matrix.

 If P is a permutation matrix then P 1  P t .

 Let A be an m  n matrix.

(a)  r , s  A (Respectively, A r , s  ) is an m  n matrix which is same as A except that the

r  th and s  th rows (respectively, columns) are interchanged, that is, it is
obtained from A by elementary row (respectively, column) operation of type I.

(b)  r    A (Respectively, A r    )is an m  n matrix which is same as A except that the

r  th row (respectively, column) is multiplied by  , that is it is obtained from A by
applying row (respectively, column) operation of type II.

(c) rs    A (respectively, A rs    ) is an m  n matrix which is same as A expect that the

r  th row (respectively, s  th column) is replaced by r  th row (respectively, s  th
column) of A plus  times the s  th row (respectively, r  th column), that is , it is
obtained from A by applying elementary row (respectively, column) operation of
type III.

Obviously, we are pre-multiplying a by elementary matrices of size m  m , and

post-multiplying A by elementary matrices of size n  n .

 Matrices A and B of the same size are said to be row equivalent if one can be
obtained from the other by elementary row operations.

 An n  n matrix A is invertible if and only if A is a finite product of elementary

matrices.

Row reduced echelon matrix: An m  n matrix is said to be a row reduced echelon

matrix if

(a) All zero rows are in the bottom position

(b) The leading entry, that is, the first nonzero entry of each nonzero row is 1.

(c) If the first t rows are nonzero, and the leading entry of each i  th row is at the
ki  th column, for i  1,..., t , then all other entries of ki  th column are zero and
ki  ki 1 , i  2,..., t .

 Row reduced echelon form for a matrix is unique.



DI PS HOU S E , 28, J i a S a r a i , H a uz K ha s, N e a r I .I. T ., N e w D e lh i - 11 00 16 17

w w w . d i p s a c a d e m y . c o m | i nf o @ d i p s a c a d e m y . c o m | P h . 8 8 - 0 0 - 2 2 - 1 3 8 6 , (0 1 1 )- 2 6 5 3 7 5 2 7