MATHEMATICS-II (Lecture-18)

The Matrix Linear Transformation

P. Danumjaya

Department of Mathematics
BITS-Pilani K.K. Birla Goa Campus

February 26, 2024

1/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 1 / 30
Coordinate Vector

Let V be a vector space with a basis

S = {v1 , v2 , · · · , vn },

and let v be any vector in V .

Since S is a basis for V and v ∈ V then there exist unique scalars

α1 , α2 , · · · , αn such that

v = α1 v1 + α2 v2 + · · · + αn vn .

2/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 2 / 30
Coordinate Vector (Contd..)
 
α1
 α2 
The column vector uS =   is called the coordinate vector of v
 
..
 . 
αn
relative to the basis S.

The scalars α1 , α2 , · · · , αn are called the coordinates of v relative to the

basis S.

3/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 3 / 30
Transition Matrix

Let the coordinate vectors of v1 , v2 , · · · , vn relative to the basis

B2 = {w1 , w2 , · · · , wn } be [v1 ]B2 , [v2 ]B2 , · · · , [vn ]B2 .

The matrix P having these vectors as columns is called the transition

matrix from the basis B1 to the basis B2 .

Thus, the transition matrix P is

P = [[v1 ]B2 [v2 ]B2 ··· [vn ]B2 ] .

4/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 4 / 30
Example 1

Find the transition matrix P from the given basis B1 to the basis B2 of R2

B1 = {(2, 3), (1, 2)} and B2 = {(1, 0), (0, 1)}.

5/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 5 / 30
Solution

We express the vectors of B1 in terms of the vectors of B2 to get the

transition matrix P.
We have

(2, 3) = 2 (1, 0) + 3 (0, 1),

(1, 2) = 1 (1, 0) + 2 (0, 1).
   
2 1
The coordinate vectors of (2, 3) and (1, 2) are and ,
3 2
respectively.
Therefore, the transition matrix
 
2 1
P= .
3 2

6/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 6 / 30
Theorem 1

Let V and W be n and m dimensional vector spaces respectively, with

bases
B1 = {v1 , v2 , · · · , vn },
and
B2 = {w1 , w2 , · · · , wm },
respectively.

Let L : V → W be a linear transformation and v be a vector in V .

7/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 7 / 30
Theorem 1 (contd...)

The coordinate matrix [v ]B1 will be a vector in Rn and the coordinate

matrix [L(v )]B2 will be a vector in Rm .

L
v L(v)

[v]B1 [L(v)]B2
A

8/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 8 / 30
Theorem 1 (contd...)

Then
[L(v )]B2 = A [v ]B1 ,
where
A = [L(v1 )B2 , L(v2 )B2 , · · · , L(vn )B2 ] .
The matrix A defines a transformation of coordinate vectors of V in the
same way as L transforms the vectors of V .

Here, A is called the matrix representation of L with respect to the bases

B1 and B2 .

9/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 9 / 30
Proof

Let
v = α1 v1 + α2 v2 + · · · + αn vn .
Using the linearity of L, we can write

L(v ) = L (α1 v1 + α2 v2 + · · · + αn vn ) ,
= α1 L(v1 ) + α2 L(v2 ) + · · · + αn L(vn ).

Let the effect of L on the basis vectors of V be

L(v1 ) = c11 w1 + c21 w2 + · · · + cm1 wm ,

L(v2 ) = c12 w1 + c22 w2 + · · · + cm2 wm ,
..
.
L(vn ) = c1n w1 + c2n w2 + · · · + cmn wm .

10/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 10 / 30
Proof (contd...)

Thus,

L(v ) = α1 (c11 w1 + c21 w2 + · · · + cm1 wm )

+ α2 (c12 w1 + c22 w2 + · · · + cm2 wm ) + · · ·
+ αn (c1n w1 + c2n w2 + · · · + cmn wm ) ,
= (α1 c11 + α2 c12 + · · · + αn c1n ) w1
+ (α1 c21 + α2 c22 + · · · + αn c2n ) w2 + · · ·
+ (α1 cm1 + α2 cm2 + · · · + αn cmn ) wm .

11/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 11 / 30
Proof (contd...)

The coordinate vector of L(v ) is

 
α1 c11 + α2 c12 + · · · + αn c1n
 α1 c21 + α2 c22 + · · · + αn c2n 
[L(v )]B2 = 
 
.. 
 . 
α1 cm1 + α2 cm2 + · · · + αn cmn
  
c11 c12 . . . c1n α1
 c21 c22 . . . c2n    α2 
 
=  ..   ..  = A [v ]B1 ,

.. ..
 . . .  . 
cm1 cm2 . . . cmn αn

where A = [L(v1 )B2 , L(v2 )B2 , · · · , L(vn )B2 ].

12/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 12 / 30
Example 2

Consider the linear transformation L : R3 → R2 defined by

L(x1 , x2 , x3 ) = (x1 + x2 , 2x3 ) .

Find the matrix of L with respect to the bases

B1 = {(1, 1, 0), (0, 1, 4), (1, 2, 3)}

and
B2 = {(1, 0), (0, 2)}.

13/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 13 / 30
Solution

Let v1 = (1, 1, 0), v2 = (0, 1, 4), v3 = (1, 2, 3) and

w1 = (1, 0), w2 = (0, 2).

We compute

L(v1 ) = L(1, 1, 0) = (2, 0) = 2(1, 0) + 0(0, 2) = 2w1 + 0w2 ,

L(v2 ) = L(0, 1, 4) = (1, 8) = 1(1, 0) + 4(0, 2) = 1w1 + 4w2 ,
L(v3 ) = L(1, 2, 3) = (3, 6) = 3(1, 0) + 3(0, 2) = 3w1 + 3w2 .

Thus, the coordinate vectors of L(v1 ), L(v2 ) and L(v3 ) are

     
2 1 3
, , and , respectively.
0 4 3

These vectors form the columns of the matrix of L.

14/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 14 / 30
Solution (contd...)

Therefore, the matrix L with respect to the bases B1 and B2 as follows:

 
2 1 3
A= .
0 4 3

15/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 15 / 30
Example 3

Let L : P2 → P3 be the linear transformation defined by

L (p(x)) = xp(x).

Find the matrix for L with respect to the standard bases

B1 = {1, x, x 2 } and B2 = {1, x, x 2 , x 3 }.

16/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 16 / 30
Solution

Given that L : P2 → P3 is the linear transformation defined by

L (p(x)) = xp(x).

We compute

L(1) = x = 0(1) + 1(x) + 0(x 2 ) + 0(x 3 ),

L(x) = x 2 = 0(1) + 0(x) + 1(x 2 ) + 0(x 3 ),
L(x 2 ) = x 3 = 0(1) + 0(x) + 0(x 2 ) + 1(x 3 ).

17/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 17 / 30
Solution (contd...)

Thus, the coordinate vectors of L(1), L(x) and L(x 2 ) are

     
0 0 0
 1   0   0 
 ,   , and   , respectively.
 0   1   0 
0 0 1

These vectors form the columns of the matrix of L.

Therefore, the matrix L with respect to the bases B1 and B2 as follows:

 
0 0 0
 1 0 0 
A= .
 0 1 0 
0 0 1

18/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 18 / 30
Example 4

Let L : P1 → P2 be defined by

L (p(t)) = t p(t) + p(0).

Let

B1 = {t + 1, t − 1}, and B2 = {t 2 + 1, t − 1, t + 1}

be bases for P1 and P2 , respectively.

Find the matrix of L with respect to B1 and B2 .

Compute L(−3t + 3) using the definition of L and also using the matrix of
L.

19/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 19 / 30
Solution

Given that L : P1 → P2 and

L (p(t)) = t p(t) + p(0).

We calculate
L(t + 1) = t (t + 1) + 1 = t 2 + t + 1.
We now find the coordinate vector of L(t + 1)

t 2 + t + 1 = α1 (t 2 + 1) + α2 (t − 1) + α3 (t + 1).

Comparing the coefficients on both sides, we find

1
α1 = 1, α2 = α3 = .
2

20/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 20 / 30
Solution (contd...)

Therefore, we have  
1
[L(t + 1)]B2 =  1/2  .
1/2

21/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 21 / 30
Solution (contd...)

We now calculate

L(t − 1) = t (t − 1) − 1 = t 2 − t − 1.

We now find the coordinate vector of L(t − 1)

t 2 − t − 1 = β1 (t 2 + 1) + β2 (t − 1) + β3 (t + 1).

Comparing the coefficients on both sides, we find

1 3
β1 = 1, β2 = , β3 = − .
2 2
Therefore, we have  
1
[L(t − 1)]B2 =  1/2  .
−3/2
22/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 22 / 30
Solution (contd...)

The matrix of L with respect to B1 and B2 is

 
1 1
A =  21 1 
2 .
1 3
2 −2

Using the definition, we find

L(−3t + 3) = t (−3t + 3) + 3 = −3t 2 + 3t + 3.

To compute L(−3t + 3) using A, we first find [(−3t + 3)]B1 .

We find that  
0
[(−3t + 3)]B1 = .
−3

23/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 23 / 30
Solution (contd...)

We note that

[L(−3t + 3)]B2 = A [(−3t + 3)]B1 ,

   
1 1   −3
0
=  12 1 
2 =  −3/2  .
1 3 −3
2 −2 9/2

Hence,

3 9
L [(−3t + 3)] = −3 t 2 + 1 − (t − 1) + (t + 1) = −3t 2 + 3t + 3.
2 2

24/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 24 / 30
Example 5

Let L : P4 → P4 be a linear transformation defined by

Z t
L (p(t)) = p ′ (s) ds,
1

for all p(t) in P4 .

Let
B1 = {1, t, t 2 , t 3 , t 4 },
and
B2 = {t − 1, t + 1, t 2 − t 4 , t 3 + t 4 , t 2 + t}
be the bases for P4 .

Find the matrix of L with respect to B1 and B2 .

25/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 25 / 30
Solution

Let L : P4 → P4 be defined by
Z t
L (p(t)) = p ′ (s) ds.
1

Let

B1 = {1, t, t 2 , t 3 , t 4 }, and B2 = {t − 1, t + 1, t 2 − t 4 , t 3 + t 4 , t 2 + t}

are the bases for P4 .

We calculate

L(1) = 0 = 0(t − 1) + 0(t + 1) + 0(t 2 − t 4 ) + 0(t 3 + t 4 ) + 0(t 2 + t).

26/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 26 / 30
Solution (contd...)

Z t
L(t) = ds = t − 1 = 1(t − 1) + 0(t + 1) + 0(t 2 − t 4 )
1
+0(t 3 + t 4 ) + 0(t 2 + t).

Z t
L(t 2 ) = 2 sds = t 2 − 1 = α1 (t − 1) + α2 (t + 1)
1
+α3 (t 2 − t 4 ) + α4 (t 3 + t 4 ) + α5 (t 2 + t).

This gives
α1 = 0, α2 = −1, α3 = 0, α4 = 0, α5 = 1.

27/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 27 / 30
Solution (contd...)

Z t
3
L(t ) = 3 s 2 ds = t 3 − 1 = β1 (t − 1) + β2 (t + 1)
1
+β3 (t 2 − t 4 ) + β4 (t 3 + t 4 ) + β5 (t 2 + t).

Solving, we obtain

β1 = 1, β2 = 0, β3 = 1, β4 = 1, β5 = −1.

Z t
L(t 4 ) = 4 s 3 ds = t 4 − 1 = γ1 (t − 1) + γ2 (t + 1)
1
+γ3 (t 2 − t 4 ) + γ4 (t 3 + t 4 ) + γ5 (t 2 + t).

28/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 28 / 30
Solution (contd...)

Solving, we obtain

γ1 = 0, γ2 = −1, γ3 = −1, γ4 = 0, γ5 = 1.

Therefore, the matrix of L with respect to B1 and B2 is

 
0 1 0 1 0
 0
 0 −1 0 −1  
A=  0 0 0 1 −1  .
 0 0 0 1 0 
0 0 1 −1 1

29/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 29 / 30
 THANK YOU...

30/30
P. Danumjaya The Matrix Linear Transformation February 26, 2024 30 / 30