Determinants

You’ve probably seen determinants of 2 × 2 matrices. The determinant is

defined inductively for any n × n matrix as follows.
Definition 1. Given A = (aij ) an n × n matrix. The determinant of A is the
number det (A) defined as follows:
1. If n = 1, det (A) = a0 .
2. If n ≥ 2,
n
X 1+j
det (A) = (−1) a1j · det (A1j )
j=1
1+n
= a11 det (A11 ) − a12 det (A12 ) + · · · + (−1) a1n det (A1n ) ,

where Aij is the sub-matrix of A obtained by deleting the i-th row and
j-th column. This is called the co-factor expansion along the first row.
Remark 2. You should familiarize yourself with the sum sign.

Example 3.
1. A = (7) ⇒ det (A) = 7
 
2 4
2. A =
−3 −5

det (A) = 2 · det (−5) − 4 · det (−3)

= 2 · (−5) − 4 · (−3) = 2
 
1 3 5
3. A = 2 1 1
3 4 2
     
1 1 2 1 2 1
det (A) = 1 · det − 3 · det + 5 · det
4 2 3 2 3 4
= 1 · (1 · 2 − 1 · 4) − 3 · (2 · 2 − 1 · 3) + 5 · (2 · 4 − 1 · 3)
= −2 − 3 + 25 = 20.

1
Remark 4. There are different notations for the determinant. For example,
   
1 3 5 1 3 5
 
A = 2 1 1 , det (A) = 2 1 1 = 20
3 4 2 3 4 2
Warning: The notation for the determinant looks similar to the notation for
the absolute value of a number, but these two concepts have nothing to do with
each other.
Theorem 5. Given an n × n matrix A, det (A) can be computed by a co-factor
expansion along any row or any column:
n
X n
X
i+j i+j
det(A) = (−1) aij · det (Aij ) = (−1) aij · det (Aij )
j=1 i=1

The first formula is called the co-factor expansion along the i-th row, and the
second formula is called the co-factor expansion along the the j-th column.
Example 6. By a co-factor expansion along the last column (checkerboard
pattern):
 
1 3 5      
 2 1 1 3 1 3
2 1 1 = 5 · 
 − 1 ·  + 2 · 

3 3 4  3 4  2 1
4 2
= 5 · (8 − 3) − (4 − 9) + 2 · (1 − 6)
= 25 + 5 − 10 = 20.

Exercise: compute this determinat by expansion along the first/second/third

row/column.


Corollary 7. Given an n × n matrix A, det AT = det (A).
Proof. The formula for det(A) developed along the first row is the formula for
det(AT ) developed along the first column.
Example 8. It is particular easy to compute det (A) when A is a triangular
matrix. For example,  
5 2 13

0
 3 11 = 5 · 3 · 2 = 20.
0 0 2

Determinants and elementary row operations

Question 9. What happens to the determinant when elementary row operations
are applied?
Theorem 10. Given an n × n matrix A.

2
 1. If a multiple of one row of A is added to another row of A to produce B,
then det (B) = det (A).
2. If two rows of A are interchanged to produce B, then det (B) = − det (A).
3. If one row of A is multiplied by λ to produce B, then det (B) = λ det (A).
Example 11. We see  
a b 

c = ad − bc.
d
1.    
c d a b 

a = cb − da = − 
b c d

2.    
λa λb a b 

c = λad − λbc = λ 
d  c d

3.
 
a + λc b + λd

 c = (a + λc) d − (b + λd) c
d 
= ad + λcd − bc − λdc
 
a b 
= ad − bc = 
 
c d

By a co-factor expansion along the last row:

 
1 2 3      
3 1 4 = 5 · 2 3 − 1 · 1 3
     1 2
1 4 3 + 2 · 

5 1 2
 4 3 1

= 5 · (8 − 3) − (4 − 9) + 2 · (1 − 6)
= 25 + 5 − 10 = 20.

Definition 12. An elementary matrix is a matrix which is obtained from

the identity matrix by one single elementary row operation.
Theorem 13. Given an n × n matrix A and E1 , . . . , Ek a sequence of n × n
elementary matrices. Then,

det (Ek Ek−1 · · · E1 A) = det (Ek ) det (Ek−1 ) · · · det (E1 ) det (A) .

Example 14. Remembering the previous examples.

   
0 1 0 1
1. E1 = is the elementary matrix R1 ↔ R2 and   = −1. Then
1 0 1 0
       
c d a b  c d a b 
= E1 ⇒  = −
 .
a b c d a b c d

3
    
λ 0 λ 0
2. E2 = is the elementary matrix R1 → λR1 and  = λ. Then
0 1 0 1
       
λa λb a b a b  a b 
= E2 ⇒  = λ  .
c d c d c d c d
   
1 1+λ 0 1
3. E3 = is the elementary matrix R1 → R1 + λR2 and  =
0 1 1 0
1. Then
       
a + λc b + λd a b a + λc b + λd a b 
= E3 ⇒  = .
c d c d c d  c d

Example 15. Consider the following matrices:

 
1 −3 9 −27  
1 1 1 0 1 3
1 
A=  , B= 1
 4 8
1 2 4 8 
0 0 3
1 3 9 27
We compute the determinant using elementary row operations:

   
1 −3 9 −27 1 −3 9 −27
 = det 0 4 −8 28 
1 1 1 1   
det 

1 2 4 8  0 5 −5 35 
1 3 9 27 0 6 0 54
 
1 −3 9 −27
0 1 −2 7 
Null () =4 · 5 · 6 · det 0

1 −1 7 


0 1 0 9
   
1 −2 7 1 −2 7
=4 · 5 · 6 · det 1 −1 7 = 4 · 5 · 6 · det 0 1 0
1 0 9 0 2 2
=2 · 4 · 5 · 6 = 120.
   
0 1 3 1 4 8
det 1 4 8 = − det 0 1 3 = −3.
0 0 3 0 0 3
Theorem 16. Given an n × n matrix A.
1. If a multiple of one column of A is added to another column of A to produce
B, then det (B) = det (A).
2. If two columns of A are interchanged to produce B, then det (B) = − det (A).
3. If one column of A is multiplied by λ to produce B, then det (B) =
λ det (A).

4
Corollary 17. Given an n × n matrix A.
1. A is invertible if and only if det (A) 6= 0.
2. Given an n × n matrix B, we have det (AB) = det (A) det (B).