MATRICES

1. INTRODUCTION :
A rectangular array of mn numbers in the form of m horizontal lines
(called rows) and n vertical lines (called columns), is called a matrix
of order m by n, written as m × n matrix.
In compact form, the matrix is represented by A=[aij]m×n.

2. SPECIAL TYPE OF MATRICES :

(a) Row Matrix (Row vector) : A = [a11, a12, ..........a1n] i.e. row
matrix has exactly one row.
é a11 ù
ê ú
ê a 21 ú
(b) Column Matrix (Column vector) : A = i.e. column
ê : ú
ê ú
ëêa m1 ûú
matrix has exactly one column.
(c) Zero or Null Matrix : (A = Om×n), An m × n matrix whose all
entries are zero.
(d) Horizontal Matrix : A matrix of order m × n is a horizontal
matrix if n > m.
(e) Vertical Matrix : A matrix of order m × n is a vertical matrix
if m > n.
(f) Square Matrix : (Order n) If number of rows = number of
column, then matrix is a square matrix.
Note :
(i) The pair of elements aij & aji are called Conjugate Elements.
(ii) The elements a11, a22, a33,....... ann are called Diagonal
Elements. The line along which the diagonal elements
lie is called ''Principal or Leading diagonal. "The quantity
å a ii = trace of the matrix written as, tr (A)
3. SQUARE MATRICES :
SQUARE MATRICES

Triangular Matrix

1 3 –2 1 0 0
A= 0 2 4 ; B = 2 –3 0
0 0 5 4 3 3

Upper Triangular Lower Triangular

aij = 0 " i > j aij = 0 " i < j

Diagonal Matrix denoted as

A = diag (a11 , a22 ........, ann)
where aij = 0 for i ¹ j

Scalar Matrix
Unit or Identity Matrix
a 0 0
0 a 0 aij= 1 if i = j
0 0 a 0 if i ¹ j

If a = a =a = a
11 22 33
If a = a =a = 1
11 22 33

Note :
(i) Minimum number of zeros in triangular matrix of order n
= n(n – 1)/2.
(ii) Minimum number of zeros in a diagonal matrix of order n
= n (n – 1).
(iii) Null square matrix is also a diagonal matrix.

4. EQUALITY OF MATRICES :
Matrices A = [aij] & B= [bij] are equal if,
(a) both have the same order.
(b) aij = bij for each pair of i & j .
5. ALGEBRA OF MATRICES :
(I) Addition : A + B = [aij + bij] where A & B are of the same
order.
(a) Addition of matrices is commutative : A + B = B + A
(b) Matrix addition is associative : (A + B) + C = A + (B + C)
(c) A + O = O + A (Additive identity)
(d) A + (–A) = (–A) + A = O (Additive inverse)

(II) Multiplication of A Matrix By A Scalar :

éa b c ù éka kb kc ù
êb c a ú ê ú
If A= ê ú , then kA = êkb kc ka ú
êë c a b úû êë kc ka kb úû

(III) Multiplication of matrices (Row by Column) :

Let A be a matrix of order m × n and B be a matrix of order p
× q then the matrix multiplication AB is possible if and only if
n = p.
Let Am × n = [aij] and Bn × p = [bij], then order of AB is m × p
n
& (AB)ij = å air brj
r =1

(IV)Properties of Matrix Multiplication :

(a) AB = O A = O or B = O (in general)

Note :
If A and B are two non-zero matrices such that AB = O,
then A and B are called the divisors of zero. If A and B are
two matrices such that
(i) AB = BA then A and B are said to commute
(ii) AB = –BA then A and B are said to anticommute

(b) Matrix Multiplication Is Associative :

If A, B & C are conformable for the product AB & BC,
then (AB)C = A(BC)
 (c) Distributivity :

A(B + C) = AB + AC ù
Provided A,B & C are conformable
(A + B)C = AC + BC úû
for respective products
(V) Positive Integral Powers of A square matrix :
(a) AmAn = Am+n
(b) (Am)n = Amn = (An)m
m
(c) I = I , m,n Î N

6. CHARACTERISTIC EQUATION :
Let A be a square matrix. Then the polynomial in x, |A – xI| is
called as characteristic polynomial of A & the equation |A – xI| = 0
is called characteristic equation of A.

7. CAYLEY - HAMILTON THEOREM :

Every square matrix A satisfy its characteristic equation
i.e. a0xn + a1xn–1 + ........ + an–1x + an = 0 is the characteristic
equation of matrix A, then a0An + a1An–1 + ........ + an–1A + anI = 0
8. TRANSPOSE OF A MATRIX : (Changing rows & columns)
Let A be any matrix of order m × n. Then transpose of A is A T or A'
of order n × m and (AT)ij = (A)ji.
Properties of transpose :
If AT & BT denote the transpose of A and B
(a) (A+B)T = AT+BT ; note that A & B have the same order.
(b) (A B)T = BT AT (Reversal law) A & B are conformable for matrix
product AB
(c) (AT)T = A
(d) (kA)T = kAT, where k is a scalar.
General : (A1. A2,....... An)T = A nT . ...... . A 2T . A1T
(reversal law for transpose)
9. ORTHOGONAL MATRIX
A square matrix is said to be orthogonal matrix if A A T = I
Note :
(i) The determinant value of orthogonal matrix is either 1 or –1.
Hence orthogonal matrix is always invertible
(ii) AAT = I = AT A Hence A–1 = AT .
10. SOME SPECIAL SQUARE MATRICES :
(a) Idempotent Matrix : A square matrix is idempotent provided
A2 = A.
For idempotent matrix note the following :
(i) An = A " n Î N.
(ii) determinant value of idempotent matrix is either 0 or 1
(iii) If idempotent matrix is invertible then it will be an identity
matrix i.e. I.
(b) Periodic Matrix : A square matrix which satisfies the relation
Ak+1 = A, for some positive integer K, is a periodic matrix. The
period of the matrix is the least value of K for which this holds
true.
Note that period of an idempotent matrix is 1.
(c) Nilpotent Matrix : A square matrix is said to be nilpotent
matrix of order m, m Î N , if Am = O, Am–1 ¹ O.
Note that a nilpotent matrix will not be invertible.
(d) Involutary Matrix : If A2 = I, the matrix is said to be an
involutary matrix.
Note that A = A –1 for an involutary matrix.
(e) If A and B are square matrices of same order and AB = BA
then
(A + B)n = nC0An + nC1An–1B + nC2An–2B2 + ........... + nCnBn
11. SYMMETRIC & SKEW SYMMETRIC MATRIX :
(a) Symmetric matrix :
For symmetric matrix A = AT i.e. aij = aji " i,j.
Note : Maximum number of distinct entries in any symmetric
n(n + 1)
matrix of order n is .
2
(b) Skew symmetric matrix :
Square matrix A = [aij] is said to be skew symmetric if AT = –A
i.e. aij = –aji " i & j. Hence if A is skew symmetric, then
a i i = -a i i Þ a i i = 0 " i .
Thus the diagonal elements of a skew square matrix are all
zero, but not the converse.
(c) Properties of symmetric & skew symmetric matrix :
(i) Let A be any square matrix then, A + AT is a symmetric
matrix & A – AT is a skew symmetric matrix.
(ii) The sum of two symmetric matrix is a symmetric matrix
and the sum of two skew symmetric matrix is a skew
symmetric matrix.
(iii) If A & B are symmetric matrices then,
(1) AB + BA is a symmetric matrix
(2) AB – BA is a skew symmetric matrix.
(iv) Every square matrix can be uniquely expressed as a sum or
difference of a symmetric and a skew symmetric matrix.

1 1
A= (A + A T ) + (A - A T )
2 4244
14 3 2 4244
14 3
symmetric skew symmetric

1 T 1
and A= (A + A) - (A T - A)
2 2
12. ADJOINT OF A SQUARE MATRIX :
æ a11 a12 a13 ö
Let A = [aij] = ç a21 a22 a23 ÷ be a square matrix and let the
ç ÷
çè a a32 a ÷ø
31 33

matrix formed by the cofactors of [ai j] in determinant |A| is

æ C11 C12 C13 ö æ C11 C21 C31 ö

ç ÷
ç C21 C22 C23 ÷ . Then (adj A) = ç C12 C22 C32 ÷
çC ç ÷
è 31 C32 C33 ÷ø çè C
13 C23 C ÷ø
33

= Transpose of cofactor matrix.

Note :
If A be a square matrix of order n, then
(i) A(adj A) = | A | In = (adj A) . A
(ii) |adj A| = |A|n–1, n > 2
(iii) adj(adj A) = |A|n–2 A, |A|¹ 0.
(iv) adj (AB) = (adj B) (adj A)
(v) adj (KA) = Kn–1 (adj A) , where K is a scalar
13. INVERSE OF A MATRIX (Reciprocal Matrix) :
A square matrix A said to be invertible (non singular) if there exists
a matrix B such that, AB = I (Note that AB = I Û BA = I)
B is called the inverse (reciprocal) of A and is denoted by A –1 . Thus
A -1 = B Û AB = I = BA
We have, A.(adj A) =|A | I n
Þ A -1 . A(adj A) = A -1 In |A|
Þ In (adj A) = A -1 |A|I n
(adj A)
\ A -1 =
|A|
Note : The necessary and sufficient condition for a square matrix A
to be invertible is that |A|¹ 0
Theorem : If A & B are invertible matrices of the same order, then
(AB) -1 = B -1 A -1 .
 Note :
(i) If A be an invertible matrix, then AT is also invertible &
(A T )-1 = (A -1 ) T .
(ii) If A is invertible ,
(a) (A -1 )-1 = A (b) (A k )-1 = (A -1 )k = A - k ; k Î N
1
(iii) |A–1| = .
|A|
14. SYSTEM OF EQUATION & CRITERIA FOR CONSISTENCY
Gauss - Jordan method :
Example :
a1x + b1y + c1z = d1
a2x + b2y + c2z = d2
a3x + b3y + c3z = d3
é a1 x + b1 y + c1 z ù é d1 ù é a1 b1 c1 ù é x ù é d1 ù
ê ú ê ú êa b2 c2 úú êy ú = ê d ú
Þ ê a2 x + b2 y + c2 z ú = d
ê ú 2 Þ ê 2 ê ú ê 2ú
êë a3 x + b3 y + c3 z úû êëd 3 úû êë a3 b3 c3 úû êë z úû êë d3 úû

Þ AX=B Þ A–1AX = A–1B, if |A| ¹ 0.

Adj A
Þ X = A–1 B = .B
|A|
Note :
(i) If |A|¹ 0 , system is consistent having unique solution
(ii) If |A|¹ 0 & (adj A) . B ¹ O(Null matrix), system is consistent
having unique non-trivial solution.
(iii) If |A|¹ 0 & (adj A) . B = O (Null matrix), system is consistent
having trivial solution.
(iv) If | A | = 0 , then matrix method fails

If (adj A) . B = O (null matrix) If (adj A) . B ¹ O

Infinite solutions Inconsistent (no solution)

or no solution