LESSON 3: Inverse of a Matrix

“The only way to learn mathematics is to do mathematics ” – Paul Halmos

O.M. “In the previous lesson, we saw that division is carried out among matrices the same way it
is done among numbers: using the concept of an inverse. In this lesson, we focus specifically on
how to find the inverse of a matrix.”

3.1 ADJOINT OF A MATRIX

adj(A) = ( 𝑑 −𝑏)
−𝑐 𝑎
Q6: What do you notice happened to the leading diagonal in adj(A)?
__________________________________________________________________________

Q7: What do you notice happened to the opposite diagonal in adj(A)?

Example 1:
3 1 𝟒 −𝟏
If B = ( ) , then adj(B) = ( )
2 4 −𝟐 𝟑

Example 2:
3 −1 𝟒 𝟏
If C = ( ) , then adj(C) = ( ) //Note, the signs of the numbers in the opposite
2 4 −𝟐 𝟑
diagonal
changes, but not their positions! //
Example 3:
5 0 𝟖 𝟎
Let D = ( ), then adj(D) = ( ) // There is no such thing as – 0. 0 is always 0, neither
6 8 −𝟔 𝟓
positive nor negative. //
3.2 CALCULATING THE INVERSE OF A MATRIX
To calculate the inverse of a matrix, we take the adjoint of that matrix and multiply it by the
inverse of its determinant. Therefore, given a matrix M, its inverse , denoted by M-1 , is given
1
by the calculation: M-1 = × adj(A). (Recall that det (A) is the determinant of matrix
det(𝐴)
A)
Note: You can only find the inverse of a square matrix!

Example 4:
3 1
Given a matrix A = ( ) , calculate A-1 , the inverse of A.
2 4
Solution:
Det(A) = (3 × 4) – (1 × 2) = 12 - 2 = 10
1
⇒ the inverse of the determinant is
10
4 −1
Adj(A) = ( )
−2 3
1 1
1 4 −1 ×4 × −1
∴ A-1 = × (
−2 3
) = (1 10 10
1 )
10 × −2 × 3
10 10

𝟐 𝟏
−
𝟓 𝟏𝟎
= ( 𝟏 𝟑 )
−
𝟓 𝟏𝟎

Example 5:
1 1
Let B = ( ) . Find B-1
3 4
Solution:
Det (B) = (1 × 4) - (3 × 1) = 4 – 3 = 1
⇒ the inverse of the determinant is 1 // Yes, this is possible! //
4 −1
Adj (B) = ( )
−3 1
4 −1 𝟒 −𝟏
∴ B-1 = 1 × (
−3 1
) =(
−𝟑 𝟏
)
3.3 THE INVERSE RULE: A × A-1 = I
1 0
Do you remember this guy? I= ( )
0 1
The equation above says: “If a matrix A is multiplied to its inverse A-1, the result will be the
identity matrix." But let’s see if this is true with an example.
1
3 1 2 −2
A= ( ) and A-1 = ( 3 ) //See example 4 for working.//
2 4 −1 2
2 1
3 1 − 10
Then A × A-1 = ( ) × ( 51 3 )
2 4 − 5 10
2 1 1 3
(3 × 5) + (1 × − 5) (3 × − 10) + (1 × 10 )
= ( 2 1 1 3
) //Verify these sub-
(2 × 5) + (4 × − 5) (2 × − 10) + (4 × 10 )

calculations.//
6 1 3 3
(5− ) (− 10 + 10
5
= ( 4 4 1 6 ) //Also verify these sub-calculations.//
(5− ) (− + )
5 5 5

5
0
= (5 5)
0 5

𝟏 𝟎
= ( )=I
𝟎 𝟏
The result A × A-1 = I can of course be proven algebraically, however the process is quite horrible
to behold so we’ll avoid it. In the meantime, trust that this is a fact of mathematics and therefore
you should absorb this into your memory. The identity matrix in the matrix world has the same
function as 1 in the number world. That is, just as any number multiplied to its multiplicative
inverse = 1, so too any matrix multiplied to its inverse will = the identity matrix. Observe the
parallel below:

Numbers Matrices

1 A × A-1 = I
N× =1
𝑁
3.4 SINGULAR MATRIX
Recall that the determinant of a matrix tells us “whether or not the matrix has an
inverse”(Lesson 2.2). Matrices that do not have an inverse are called singular. A matrix will not
1
have an inverse if its determinant is 0. Why? Because it would not be possible to compute
0
(Division by zero in mathematics is not computable and at best ∞ is given as the result). So a
singular matrix is a square matrix whose determinant is 0 and therefore does not have an inverse.
For example,
2 6
If say S = ( ) , then det (S) = (2 × 3) – (6 × 1) = 6 – 6 = 0. Since det(S) = 0, S is a singular
1 3
matrix.

TAKE-AWAYS
𝑎 𝑏 𝑑 −𝑏
 The adjoint of a matrix A = ( ), denoted as adj(A), = ( )
𝑐 𝑑 −𝑐 𝑎
 Only square matrices have an adjoint.
1
 A-1, the inverse of a matrix A, is given by × adj(A)
det 𝐴
 A × A-1 = I that is, any matrix multiplied to its inverse will result in the identity matrix.
 A singular matrix is a square matrix whose determinant is 0 and therefore does not have
an inverse.