DETERMINANT

By
Sonia Young Tan
June, 2020

S.Y.Tan
 DETERMINANT OF
SQUARE MATRIX
Definition: A permutation on the set Sn = { 1, 2, . . . , n } of
the first n positive integers is rearrangement of the n
elements of the set Sn.
For S1 = { 1} , there is 1! = 1 permutation possible.

For S2 = { 1,2} , there is 2! = 2 permutations possible.

1 2 and 2 1
For S3 = { 1, 2, 3} , there is 3! = 6 permutations
possible.
123 132 213 231 312 321
In general, a set Sn of n elements has n!
permutations possible for its n elements.
S.Y.Tan
 DETERMINANT OF
SQUARE MATRIX
Definition: A permutation p1 p2 … pn on the set Sn of the
first n positive integers is said to have an inversion if a
larger number precedes a smaller one.
Example 1 1 2 3 no inversion
Example 2 4 1 2 3 5 has 3 inversions.

Example 3 4 3 5 1 2 has 7 inversions.

S.Y.Tan
 DETERMINANT OF
SQUARE MATRIX
Definition: A permutation p1 p2 … pn is said to be even if
the total number of inversions is even.
Definition: A permutation p1 p2 … pn is said to be odd if the
total number of inversions is odd.

Example 1 1 2 3 no inversion or 0 inversion (even)

Example 2 4 1 2 3 5 has 3 inversions (odd)

Example 3 4 3 5 1 2 has 7 inversions (odd)

S.Y.Tan
 DETERMINANT OF
SQUARE MATRIX
In general, there are (n!/2) even permutations and
(n!/2) odd permutations.

For S3 = { 1, 2, 3} , there is 3! = 6 permutations possible.

1 2 3 0 inversion (even) 1 3 2 1 inversion (odd)

2 1 3 1 inversion (odd) 2 3 1 2 inversions (even)
3 1 2 2 inversions (even) 3 2 1 3 inversions (odd)

For S3 = { 1, 2, 3} , there is 3!/2 = 6/2 = 3

even permutations and 3 odd permutations.

S.Y.Tan
 DETERMINANT OF
SQUARE MATRIX
Definition: The determinant of a square matrix A of order n,
denoted by lAl is defined as

A =  ()a 1p1 a 2 p 2 ...a np n

where p1p 2 ... p n is a permutation on Sn = {1, 2, 3, …, n}

and take (+) if permutation is even , take (-) if permutation is
odd and the summation ranges over all permutations on Sn .

S.Y.Tan
 I. 2 X 2 matrix
 a11 a12 
Given a 2 x 2 matrix A= 
 21
a a 22 

the determinant of A is given by

A =  ()a1p1 a 2p2 = a11a 22 − a12a 21
where p1p 2 is possible permutation on S2

That is,
a11 a12
A = = a11a22 – a12a21
a 21 a 22

S.Y.Tan
 2 3
= 2(1) – 3 (-4) = 2 + 12 = 14
−4 1

1 2
= 1(0) – 2 (5) = 0 - 10 = - 10
5 0

S.Y.Tan
 II. 3 x 3 matrix
 a11 a12 a13 
Given a 3 x 3 matrix A = a 21 a 22 a 23 
 a 31 a 32 a 33 

A =  () a1p1 a 2p2 a 3p3 = a11a 22a 33 − a11a 23a 32 − a12a 21a 33
+ a12a 23a 31 + a13a 21a 32 − a13a 22a 31
where p1p 2 p3 is possible permutation on S3 .

S.Y.Tan
A =  ()a1p1 a 2p2 a 3p3 = a11a 22a 33 − a11a 23a 32 − a12a 21a 33
+ a12a 23a 31 + a13a 21a 32 − a13a 22a 31

a11 a12 a13 a11 a12 a13

A = a 21 a 22 a 23 a 21 a 22 a 23
a 31 a 32 a 33 a 31 a 32 a 33
= a a22a33 + a12a23a + a13a21a32
11 31

- a11a23a32 - a12a21a33 - a13a22a31

This method is known as the Sarrus’s Rule or
sometimes we called it the Criss-Cross method
Sarrus’s Rule or Criss-Cross method works
ONLY on 2x2 and 3x3 matrices. S.Y.Tan
 2 4 3
A = 4 1 2 
5 2 1

2 4 3 2 4 3
A = 4 1 2 4 1 2
5 2 1 5 2 1

= 2(1)(1)+ 4(2)(5)+ 3(4)(2) -2(2)(2) -4(4)(1) -3(1)(5)

= 2 + 40 + 24 – 8 - 16 -15 = 27

S.Y.Tan
Co-factor Expansion Method
a11 a12 a13
A = a 21 a 22 a 23
a 31 a 32 a 33

A = a11a22a33 +a12a23a31 +a13a21a32

-a11a23a32 - a12a21a33 - a13a22a31
=a11a22a33 -a11a23a32 + a12a23a31 - a12a21a 33

+ a a21a32 -a13a22a31
13

= a11(a22a33-a23a32) - a12(a21a33-a23a31)
+ a13(a21a32-a22a31)
a 22 a 23 a 21 a 23 a 21 a 22
=a 11(-1)
1+1 +a 12(-1)
1+2
+ a13(-1)1+3
a 32 a 33 a 31 a 33 a 31 a 32

Cofactor expansion along 1st row

S.Y.Tan
 Cofactor expansion along 1st row
a11 a12 a13
A = a 21 a 22 a 23
a 31 a 32 a 33
a 22 a 23 a 21 a 23 a 21 a 22
= a11(-1)1+1 a 32 a 33
+ a 12(-1)1+2 + a13(-1)1+3 a a 32
a 31 a 33 31

Cofactor of a11 Cofactor of a12 Cofactor of a13

C12 C13
C11

S.Y.Tan
 Cofactor expansion along 1st row
2 4 3
A = 4 1 2
5 2 1
1 2 4 2 4 1
= 2(-1)1+1 + 4(-1) 1+2 + 3(-1)1+3 5 2
2 1 5 1

= 2(1 – 4) - 4 (4 - 10) + 3 (8 - 5)
= 2(– 3) - 4 (-6) + 3 (3)
= – 6 + 24 + 9 = 27

S.Y.Tan
 Cofactor expansion along 2nd colunm
2 4 3
A = 4 1 2
5 2 1
4 2 2 3 2 3
= 4(-1)1+2 + 1(-1) 2+2 + 2(-1)3+2 4 2
5 1 5 1

= -4(4 –10) + 1 (2 - 15) - 2 (4 - 12)

= -4(– 6) + 1 (-13) -2 (- 8)
= 24 - 13 +16 = 27

S.Y.Tan
Evaluating Determinant Using Cofactor
Expansion Method
Let A be n x n matrix. Let Mij be
the submatrix of A obtained by
deleting the ith row and jth
column of A. The minor of an
element aij is the determinant
of Mij, lMijl. The cofactor of an
element aij, denoted by cij, is
cij = (-1)i+j lMijl.

S.Y.Tan
Evaluating Determinant Using Cofactor
Expansion Method
(Can be done along any row or any column)

Let A be n x n matrix.
Expanding along the ith row, we have

A = a i1ci1 + a i 2ci 2 + a i 3ci 3 + ... + a in cin

A = Sum of products obtained by

multiplying entries of ith row with
their corresponding cofactors

S.Y.Tan
 Evaluating Determinant Using Cofactor
Expansion Method
Let A be n x n matrix.
Expanding along the jth column, we have
A = a1jc1j + a 2 jc 2 j + a 3 jc 3 j + ... + anjcnj
A = Sum of products obtained by
multiplying entries of jth column with
their corresponding cofactors
Tip: Choose the row or column with
the most number of zeros and
evaluate determinant by cofactor
expansion method along that row
or column. S.Y.Tan
Determinant of a square matrix of order n > 3
we use cofactor expansion method
Expand along 2nd column as it has the most number of zeros.

1 2 −3 4 −5 4 −1
−5 0 4 −1 = -2 3 0 −3
A =
3 0 0 −3 2 −2 3
2 0 −2 3 Expand along 2nd row

= -2 [ - 3 (4(3) - (-1)(-2)) - (-3) ((-5)(-2) - (4)(2)) ]

= -2 [ - 3 (12 - 2) + 3 (10 - 8) ]
= -2 [ - 3 (10) + 3 (2) ] = -2 [ - 30 + 6 ]
= -2 [ - 24 ]= 48

S.Y.Tan
Theorem: Let A be n x n matrix,

 A if i = k
a i1ck1 + a i 2ck 2 + ... + a in ckn =
 0 if i  k
or
 A if j = k
a1jc1k + a 2 jc 2k + ... + anjcnk =
 0 if j  k
i.e. Sum of products of entries of a
row (column) with cofactors of
another row (column) is zero.

S.Y.Tan
Properties of Determinants
1) For square matrix A of order n, lATl= lAl
1 2
A = = 1(4) − 2(3) = 4 − 6 = −2
3 4
1 3
A T
= = 1(4) − 3(2) = 4 − 6 = −2
2 4
2) If a square matrix A of order n has a
row or column of zeros, lAl = 0.
−5 4 −1
3 0 −3 = 0
0 0 0
Expanding along 3rd row where all entries are zeros and
these 0s multiply with their corresponding cofactors will
result to 0s and sum of these products will also be 0.
S.Y.Tan
Properties of Determinants
3) If B is obtained from A by interchanging
2 rows (columns) of A then lBl= - lAl .
1 2
A = = 1(3) − 2( −5) = 3 + 10 = 13
−5 3

Apply EROI (R 1  R 2 ) to get B

−5 3
B = = ( −5)2 − 3(1)− = −10 − 3 = −13
1 2

−5 3 1 2
B = =− = −13 = − A
1 2 −5 3
S.Y.Tan
Properties of Determinants
4) If B is obtained from A by multiplying a
row (column) of A by a nonzero constant k
then lBl= k lAl .

1 2
A = = 1(3) − 2( −5) = 3 + 10 = 13
−5 3

Apply EROII (2R 1 ) to get B

2 4
B= = 2(3) − 4( −5)− = 6 + 20 = 26 = 2(13) = 2 A
−5 3

Note: Common factor from a row or

column can be factored out of a
determinant. S.Y.Tan
Properties of Determinants
Note: Common factor from a row or column
can be factored out of a determinant.
Common factor in 1st row Common factor in 2nd column

2 4 6 1 2 3 1 1 3
A = 1 2 5 =21 2 5 = 2(2) 1 1 5
7 8 6 7 8 6 7 4 6
A = 4 (1[6 − 20] − 1[6 − 35] + 3[4 − 7])
= 4( −14 + 29 − 9) = 4(6) = 24
Note: kA  k A
5) For square matrix A of order n,
lkAl = kn lAl.
S.Y.Tan
Properties of Determinants
6) If B is obtained from A by adding multiple
of a row (column) to another row (column) of
A then lBl = lAl .
1 2 3 1 2 3
A= 
1 2 5
 ( −1) R1 + R 2
0
 0 2
 =B

7 8 6
 
7 8 6

A = 1[12 − 40] − 2[6 − 35] + 3[8 − 14]
= ( −28 + 58 − 18) = 12
B = −0 + 0 − 2[8 − 14] = ( −2)( −6) = 12
( −1)R 1 + R 2

1 2 3 1 2 3
A = 1 2 5 = 0 0 2 = B = 12
7 8 6 7 8 6 S.Y.Tan
Properties of Determinants
7) If A has two identical rows (columns) then
lAl = 0.
( −1)R 1 + R 3
1 2 3 1 2 3
A = 1 2 5 = 1 2 5 = B =0
1 2 3 0 0 0

8) If A and B are square matrices of order n,

then l A B l = lAl lBl.
1 2
A = = 4 − 6 = −2
3 4
−1 3
B = = −2 − 15 = −17
5 2
S.Y.Tan
Properties of Determinants
8) If A and B are square matrices of order n,
then l A B l = lAl lBl.
 1 2   −1 3   −1 + 10 3 + 4   9 7 
AB =     =  = 
 3 4  5 2   − 3 + 20 9 + 8   17 17 
9 7
AB = = 9(17) − 7(17) = 153 − 119 = 34
17 17
1 2 −1 3
A = = 4 − 6 = −2 B = = −2 − 15 = −17
3 4 5 2
AB = 34 = ( −2)( −17) = A B

* It follows that for square matrix A,

lAn l=lAln. S.Y.Tan
Properties of Determinants
9) The determinant of an upper triangular or a
lower triangular matrix A is the product of
its main diagonal entries.
Expand along the 1st row
1 0 0
B = 2 3 0 = 1(3(6) − 0(5)) = 1(18 − 0) = 18 = 1(3)(6)
4 5 6
Expand along the 1st column
1 2 3
A = 0 4 5 = 1(4(6) − 5(0)) = 1(24) = 24 = 1(4)(6)
0 0 6

S.Y.Tan
Properties of Determinants
All these properties of determinant can be utilized
in simplifying the evaluation of determinant of a
square matrix A especially those with higher order.
Common factor in 3rd row R1  R 3

4 3 2 4 3 2 1 2 3
A = 3 −2 5 =2 3 −2 5 = −2 3 −2 5
2 4 6 1 2 3 4 3 2
−3R 1 + R 2
R2  R3 Common factor in 2nd row
−4R 1 + R 3
1 2 3 1 2 3 1 2 3
A = −2 0 −8 −4 = − ( −2) 0 −5 −10 = 2( −5) 0 1 2
0 −5 −10 0 −8 −4 0 −8 −4
8R 2 + R 3 Determinant of upper
triangular matrix
1 2 3 1 2 3
A = −10 0 1 2 = −10 0 1 2 = ( −10)(1)(1)(12) = −120
0 −8 −4 0 0 12 S.Y.Tan
Properties of Determinants
− R1 + R 2 Expand along 2nd column

1 2 −3 4 1 2 −3 4
−5 4 −1
−4 2 1 3 −5 0 4 −1
B = = = ( −2) 3 0 −3
3 0 0 −3 3 0 0 −3
2 −2 3
2 0 −2 3 2 0 −2 3

C1 + C3 Expand along 2nd row

−5 4 −1 −5 4 −6
4 −6
B = ( −2) 3 0 −3 = ( −2) 3 0 0 = ( −2)( −3)
−2 5
2 −2 3 2 −2 5

4 −6
B =6 = 6(20 − 12) = 6(8) = 48
−2 5

S.Y.Tan
Theorem If A is an nxn matrix then A is
non-singular if and only if lAl  0.

Equivalently, If lAl=0 then A is singular.

That is A-1 does not exist.

Corollary If A is an nxn matrix then rank of

A is n if and only if lAl  0.

Corollary If A is an nxn matrix then AX=O

has non-trivial solution if and only if lAl = 0.

S.Y.Tan
 The adjoint of square matrix A, denoted by
Adj A, is defined as
T
 c11 c12 ... c1n   c11 c 21 ... cn1 
c 
c 22 ... c 2n   c 12 c 22 
... cn2 
AdjA =  21
= 
   
   
 cn1 cn2 ... cnn   c1n c 2n ... cnn 
where cij 's are cofactors of a ij 's.

−1 1
Theorem :If A is non − sin gular then A = AdjA.
A

S.Y.Tan
Getting A-1 of 2x2 matrix using adjoint of A

a b
A=
c d 
c11 = (-1)1+1 d = d c21 = (-1)2+1 b = -b

c12 = (-1)1+2 c =-c c22 = (-1)2+2 a = a

 a b   d − b
A=   AdjA = − c 
 c d   a

−1 1 1  d − b
A = AdjA =  
A A  −c a 
S.Y.Tan
Theorem If A is an nxn matrix then A is
non-singular if and only if lAl  0.

Equivalently, If lAl=0 then A is singular.

That is A-1 does not exist.

a b
For 2x2 matrix A =   , if A is non-singular
c d

−1 1 1  d − b
A = AdjA =  
A A  −c a 

S.Y.Tan
 3 −6
A = = 15 − 12 = 3  0
−2 5
so A is non-singular and A-1 exists .

 3 −6   5 6
A=   AdjA =  
 −2 5  2 3

−1 1 1 5 6   53 2 −1
A = AdjA =   = 2  = A
A 3 2 3  3 1

S.Y.Tan
  4 − 1  0 1
B=   AdjB =  
 2 0  −2 4

4 −1
B = = 4(0) − ( −1)(2) = 0 + 2 = 2  0
2 0

−1 1 1 0 1   0 12  −1
B = AdjB =   =  −1 2  = B
B 2  −2 4   

S.Y.Tan
 Ex Find the inverse of
 2 0 −1 
A =  −1 2 1 
2 0 −1  3 −2 −4 
A = −1 2 1 = 2[−8 − ( −2)] − 0 + ( −1)[2 − 6] = −12 − 0 + 4 = −8  0
3 −2 −4
0 −1
c11 =
2 1
(-1) −2 −4
1+1 c21 = (-1)2+1 0 −1
C31 = (-1)3+1
−2 −4 2 1

−1 1
c12 = (-1)1+2 3 −4
c22 = (-1)2+2 2 −1
C32 = (-1)3+2 2 −1
−1 1
3 −4

c13 = (-1)1+3 −1 2
c23 = (-1)2+3 2 0
C33 = (-1)3+3
2 0
−1 2
3 −2 3 −2

S.Y.Tan
 Ex Find the inverse of
 2 0 −1 
A =  −1 2 1 
2 0 −1  3 −2 −4 
A = −1 2 1 = 2[−8 − ( −2)] − 0 + ( −1)[2 − 6] = −12 − 0 + 4 = −8  0
3 −2 −4
c11 =(+)(-8-(-2))=-6 c21 =(-)(0-2)= 2 C31 =(+)(0-(-2)) = 2

c12 =(-)(4-3)=-1 c22 =(+)(-8-(-3)) = -5 C32 =(-)(2-1) = -1

c13 =(+)(2-6)=-4 c23 =(-)(-4-0) = 4 C33 =(+)(4-0) = 4

S.Y.Tan
 Ex Find the inverse of
2 0 −1  −6 2 2 
A = −1 2 1 = −8  0 AdjA =  −1 −5 −1
3 −2 −4  −4 4 4 

c11 =(+)(-8-(-2))=-6 c21 =(-)(0-2)= 2 C31 =(+)(0-(-2)) = 2

c12 =(-)(4-3)=-1 c22 =(+)(-8-(-3)) = -5 C32 =(-)(2-1) = -1

c13 =(+)(2-6)=-4 c23 =(-)(-4-0) = 4 C33 =(+)(4-0) = 4

 3 −1 −1 
 4 4 4 
 −6 2 2   
1 
−1 
1 1 5 1 
A −1 = AdjA =  −1 −5  =
A −8  8 8 8 

 −4 4 4 
  
 1 −1 −1 

 2 2 2 
S.Y.Tan