Chapter Five: Linear Transformation

2.1 Linear Transformations

Definition: Let and be vector spaces over the same field . A function
is called a linear transformation (or a linear mapping) of into , if the following
conditions are satisfied:
i) ( ) ( ) ( )
ii) ( ) ( )for all and .
Example1:Define by ( ) ( ) for all . Then is a linear transformation
as:
i) ( ) ( ( )) ( )
( ) ( ) ( ) ( )
1 Let a function be given by ( ) ( ). Then show that
is a linear transformation.
Solution: Let ( )( ) and . Then
i. ( ) ( ) (( )
( )
( )
( ) ( )
( ) ( )
ii) ( ( )) ( )
=( )
= ( ) ( )
Examples:
1. Define by ( ) ( ) for all . Show that is a linear
transformation.
2. Define by ( ) ( ). Show that is a linear
transformation.
3. Define by ( ) ( ). Show that is a linear
transformation.

1|Page
 4. Define by ( ) ( ). Show that is not a linear
transformation.
5. Define a map by ( ) ̅.
a. Is is a linear operator over the real vector space ?
b. Is is a linear operator over the complex vector space ?
Solution:
4. Let ( ) and . Then ( ) ( ) and ( ) ( ). Thus
( ) ( ) and hence is not a linear transformation.
5. Since ( ) ̅̅̅̅̅̅̅ ̅ ̅ ( ) ( ) and ( ) ̅̅̅ ̅ ( ), is a
linear transformation over the real vector space. On the other hand, for , ( ) ̅̅̅
̅ ̅ ( ). Therefore, is not a linear transformation over the complex vector space.
Exercise: Linear transformation, polynomial to matrices and polynomial to polynomial
1. Define by

( ) 0 1.

Show that is a linear transformation.

2. Define by ( ( )) ( ) ( ). Show that is a linear transformation.
The following two examples are trivial.
1. Let be two vector spaces. Defined as ( ) for all .
Then is a linear transformation is called the zero transformation.
2. Let be a vector space. Defined as ( ) for all .
Then is a linear transformation, to be called the identity transformationof .
2.2 Properties of linear transformation
Theorem: Let and be two vector spaces. Suppose is a linear transformation.
Then
1. ( ) .
2. ( ) ( )for all .
3. ( ) ( ) ( )for all .
4. If , then
( ) ( ) ( ) ( ) ( )
That is, (∑ ) ∑ ( ).

2|Page
Proof:
1. By the definition of linear transformation, we have
( ) ( ) ( ) .
2. Since ( ) and is a linear transformation, we have
( ) (( ) ) ( ) ( ) ( ).
3. ( ) ( ( )) ( ) ( ) ( ) ( ).
4. To prove (4), we use induction, on .
For , we have ( ) ( ).
For , by the definition of linear transformation, we have
( ) ( ) ( ) ( ) ( ).
Now, we assume that the formula (4) is valid for n-1 vectors. That is,
( ) ( ) ( ) ( )).
Now,
( ) (( ) )
= ( ) ( ), since is a linear mapping
= ( ) ( ) ( ) ( ) by the assumption
So, the proof is completed.
Theorem: Let and be vector spaces over the same field . A function is a linear
transformation if and only if
( ) ( ) ( )for all and .
Proof: Let is a linear transformation. Since V is a vector space, for all
and .
Now,
T( ) = T( ) ( ) by the definition of linear mapping
= T( ) ( ) by the definition of linear mapping
Conversely, suppose that
( ) ( ) ( )for all and .
Now we proceed to show that is a linear transformation.
i) ( ( ) ( )
= ( ) ( )by the assumption

3|Page
 ii) T( ) ( )
( ) ( ) (by the assumption)
= ( )
= ( )
Hence T is a linear transformation.
Proposition: Let be a linear transformation over the same field. Prove that the vectors
are linearly independent whenever ( ) ( ) ( ) are linearly
independent in .
Proof: Let such that . Then
( ) ( ) . Since is a linear mapping, we have ( )
( ) ( ) . Since ( ) ( ) ( ) are linearly independent, we get
that . This shows that are linearly independent.
2.3 Linear transformations and bases
Theorem: Let and be vector spaces over the field and let * + be a basis of .
If * + is a set of arbitrary vectors in , then there exists a unique linear
transformation
such that ( ) , for .
Proof: Let * + be a basis of V and * + be any set of n-vectors in W. Then
there exist unique scalars such that
∑ .
Then
i) Define by
( ) ( ) .
That is, ( ) (∑ ) ∑ .
Thus every element of is mapped to only one element of W as the scalars ’s are unique.
Since is a vector space, every linear combination of vectors in also in . Thus∑
. So is a function from into . Moreover, since
, we have ( ) for each .
ii) Now we show that T is a linear transformation. Let . Then ∑ and
∑ for some unique scalars ,….., and in .

4|Page
 a. ( ) (∑ ∑ )
(∑ ( ) )
∑ ( ) by the definition of linear mapping.
∑ ∑
(∑ ) (∑ )
( ) ( )
b. ( ) ( ∑ )
(∑ ( ) )
∑ ( ) by the definition
∑
( ∑ ) ( )
Hence is a linear transformation.
iii) Finally, we prove that F is unique. Suppose is a linear transformation such that
( ) for each * +.
Let be any vector in . Then ∑ for some unique
scalars .Now,
( ) ∑ )
=∑ ( )since G is a linear transformation

=∑ since ( ) =, for each

= (∑ ) by the definition of
= ( )
This shows that ( ) ( )for any . Thus is unique.
Remark
1. The vectors in the above theorem are completely arbitrary; they may be
linearly dependent, independent or they may even be equal with that of the number of
basis vectors of V.
2. In determining the liner transformation from in to the assumption that
* + is basis of is essential.
Theorem: Let be a linear transformation and * + be an order base
of . Then is uniquely determined by ( ) ( ) ( ).

5|Page
Examples:
1. Let be a linear transformation. If ( ) ( )and ( ) ( ). Then
find ( ) for any .
2. Does there exist a linear transformation such that (( )) ( ) and
(( )) ( )?
3. Does there exist a linear transformation such that (( )) ( ) and
(( )) ( ) and (( )) ( )?
Solution:
,
1. Since*( )( )+ is a basis for by the above theorem we can find a unique linear
transformation satisfies the given conditions. Let ( ) . Then ( )can be
expressed as a linear combination of ( )and ( ). Now,
( ) ( ) ( )for some .

This implies and . Thus ( ) ( ) ( ).

Now,

( ) ( ( ) ( ))

( ) ( )since is a linear transformation

( ) ( )

. /

2.4 Kernel and Images

Definition: Let be a linear transformation.
1. The set of elements such that ( ) is called the kernel of or the null space
of and denoted by kernel . That is,
* ( ) +.
2. The set of elements in such that there exist in such that ( ) is called the
image or the rage of and denoted by Image . That is,
* ( ) +.
Examples: Let be defined by ( ) ( )
a. Show that is a linear transformation.

6|Page
 b. Find .
c. Find .
Solution
a. Clearly is a linear transformation.
b. *( ) ( ) ( )+
*( ) ( ) ( )+
*( ) +
*( ) +
*( ) +
* ( ) +
So kernel T is the subspace of generated by ( ).
c. *( ) ( ) ( ) for some ( ) +
*( ) ( ) ( )+
*( ) +
*( ) +
*( ) ( ) +
* ( ) ( ) }
* ( ) ( ) +
Thus Image is the subspace of R2 generated by ( ) ( ).
Theorem: Let be a linear transformation from a vector space in to over the same field .
Then
1. Kernel is a subspace of .
2. Image is a subspace of .
Proof: (1)
i. Since ( ) , we get that kernel .Thus .
ii. Let Then ( ) and ( ) . Then
( ) ( ) ( ) . Thus .
iii. Let and . Then ( ) ( ) .
(2) (i) Since ( ) , .

7|Page
 (ii) Let . Then there exist such that ( ) and ( ) .
Since is a vector space, and
( ) ( ) ( ) . This shows that .
(iii) Let and . Then there exists such that ( ) . Since vector
space, . Moreover, ( ) .
Thus is a subspace of .
Theorem: Let be a linear transformation. Then is one to one if and only if
* +.
Proof: Suppose is one to one. We need to show that contains only the zero vector. Let
. Then ( ) . Since ( ) and is one to one, we have that .
Thus * +.
Conversely, suppose * + Now we need to show is one to one. Let such
that ( ) ( ). Then ( ) ( ) . This implies ( ) . This shows that
. Since * +, we get that . Thus is one to one.
Theorem: (The rank-nullity theorem)
Let vector spaces over and let be a linear transformation. If is finite
dimensional vector space, then
( ) ( ) ( ).
2.5 Algebra of linear transformations
Notation: We denote the set of all linear transformation from a vector space into over a
field by ( ).
Let ( ) and Then
i. The sum and is a function from into defined by
( )( ) ( ) ( )for all .
ii. The scalar multiple of by , is a function from into defied by
( )( ) ( ( ))for all
Using the above fact, we can easily show that and are linear transformation from a
vector into a vector space .
Theorem: Let and be vector spaces over the filed . If and are linear transformation
from into and from into respectively, then the compose function defined by

8|Page
( )( ) ( ( ))for all is a linear transformation.
Proof: Let and . Then
i) ( ) ( ( ))
( ( ) ( )), since is a linear transformation
( ( )) ( ( )), since is a linear transformation
( ) ( )

ii) ( ) ( ( )) ( ( )) . ( ( ))/ ( )

Hence is a linear transformation.

Definition: A linear transformation is said to be invertible if there exist a linear
transformation such that is the identity function on and is the identity
function on . In this case, we write . Furthermore, is invertible if and only if is
one to one and onto.
2.6 Matrix representation of a linear transformation
2.6.1 Linear transformation associated with matrix.
Definition: For any matrix ( ) over a field . The mapping

defined by ( ) for all

is called the linear transformation associated with the matrix .
Note: In the above definition, each element of should be considered as column vector,
otherwise is not defined.
Example.

(1) Let . /. Then is an identity linear transformation.

Solution: Let .Then . /.

Now, ( ) . /. / . / . This shows that is an identity linear transformation.

(2) Let ( ). Define by

. / ( ). / ( ).

Then is a linear transformation associated with matrix .

9|Page
Solution: Let . / . / and . Then

i. ( ) (. / . /)

(. /)

( ). /

( )

( ) ( )

( ). / ( ). /

( ) ( )

ii. ( ) (. /)

( ). /

( )

( ). /

( )
Hence is a linear transformation associated with the matrix B.
In view of Definition2.6.1, we can study system of linear equations with the help of linear
transformation associated with the coefficient matrix of the system.

10 | P a g e
Consider the system where isan real matrix. Then if and only if
( ) , where is the linear transformation associated with . Thus the system has
a solution if and only if is in the range of .
 If there is exactly one element whose image is under , then the system
has exactly one solution. But if has more than one pre-image under , then the
system has exactly one solution.
 If there is no such that ( ) , (i.e., is not in the range of ), the system
has no solution.
2.6.2 The matrix of linear transformation
In section 2.6.1, we have seen that, associated to any matrix , there is a linear transformation
defined by ( ) .
In this section, we show that every linear transformation between two finite dimensional vector
spaces can be represented by a matrix.
First, let us recall about the coordinates of an element of a finite dimensional vector space with
respect to a given ordered basis of .
Suppose * + is an ordered basis for a finite dimensional vector space over a
field and . The coordinates of relative to the basis are the scalars in such
that
.
The coordinates of relative to the basis denoted by , - and

, - ( )

Let and be the finite dimensional vector spaces over with dimensions and ,
respectively.
Let * + and * + ordered bases of and respectively.
If is a liner transformation, then for any , ( )can be expressed as a liner
combination of elements of the basis . So we have,
( )
( )
( )

11 | P a g e
 , ( )- ( ) , ( )- ( ) , ( )- ( )

The coordinates ( ) with respect to the basis ,

, ( )- [ ]( ) , - , where

[ ] ,, ( )- , ( )- , ( )- -

Let and be finite dimensional vector spaces over with dimensions and , respectively.
Let be a linear transformation. Also, let and be ordered bases of and ,
respectively. Then there exists a unique matrix
( )denoted , -, with ,, ( )- , ( )- , ( )- - such
Definition (Matrix of a linear transformation)
Let and be finitedimensional vector spaces over with dimensions and , respectively.
Let be a linear transformation. Then the matrix , - is called the matrix of the
linear transformation with respect to the ordered bases and .
Note: If and are standard bases of and respectively, then the matrix , - iscalled
the standard matrix for the linear transformation .
Example: Define the linear transformation by
( ) ( ). Then
i. Find the standard matrix for .
ii. Find the matrix of relative to the ordered bases
*( )( )( )+ and *( )( )+.
Solution:
i. Since *( )( )( )+ and *( )( )+ are the standard bases of
and respectively, then

,( )- . / ,( )- . / ,( )- . /.

This shows that , - . /. Thus . / is the standard matrix of .

12 | P a g e
ii. ( ) ( ) ( ) ( ) ,( )- . /.

Similarly, ,( )- . /and ,( )- . /. The matrix relative to and is

. /. Therefore, , - . /.

13 | P a g e