PAPER

ELEMENTARY LINEAR ALGEBRA

LINEAR TRANFORMATION MATRICES

Lecture: Dr. Hmidah Nasution, M.Si

Arranged by Group 2 :

Ilham Adrianto (4222411021)

Helni W Nainggolan (4223111048)

Nazwa S Harahap (4221111005)

MATHEMATIC EDUCATION STUDY

PROGRAM FACULTY OF MATHEMATIC AND
NATURAL SCIENCE UNIVERSITAS NEGERI
MEDAN
2024
A. Kernel and Range
In this section, we will discuss the kernel and range. The definitions of kernel and range are as
follows.
Definition 6.4. If 𝑇: 𝑉 → 𝑊 is a linear transformation, then the set V that T maps to 0 ⃗ ∈ 𝑊 is
called the kernel (null space or center) of T. This set is denoted by 𝑘𝑒𝑟(𝑇), which is
ker(𝑇) = {𝑣 ∈ 𝑉; 𝑇(𝑣 ) = ⃗0}
Definition 6.5. If 𝑇: 𝑉 → 𝑊 is a linear transformation, then the set W that is an image (map) under
the transformation T is called the range (image). It is denoted by R(T). In this case,
R(𝑇) = {𝑤⃗⃗ ∈ 𝑊; 𝑤 ⃗⃗ = 𝑇(𝑣) 𝑓𝑜𝑟 𝑠𝑜𝑚𝑒 𝑣 ∈ 𝑉}
Example 1:
Suppose 𝑇: 𝑃2 → ℝ2 is defined by
𝑎−𝑏
𝑇(𝑎 + 𝑏𝑥 + 𝑐𝑥 2 ) = ( )
𝑎−𝑐
Which one represents ker (𝑇)?
a. 𝑠1 = 1 + 𝑥 + 𝑥 2
b. 𝑠2 = 1 + 2𝑥 + 𝑥 2

Solution:
a. 𝑇(𝑠1 ) = 𝑇(1 + 𝑥 + 𝑥 2 ); 𝑎 = 1, 𝑏 = 1, 𝑐 = 1
1−1 0 ⃗
=( )=( )=0
1−1 0
So, 𝑠1 ∈ ker(𝑇)
b. 𝑇(𝑠2 ) = 𝑇(1 + 2𝑥 + 𝑥 2 ); 𝑎 = 1, 𝑏 = 2, 𝑐 = 1
1−2 −1 ⃗
=( )=( )≠0
1−1 0
So, 𝑠1 ∉ ker(𝑇)

Example 2:
Suppose 𝑇: ℝ2 → ℝ2 express by formula
𝑥 2𝑥 − 𝑦
𝑇 (𝑦) = ( )
−8𝑥 + 4𝑦
Which of the following vectors is 𝑘𝑒𝑟(𝑇)?
5 3 1
a. ( ) b. ( ) c. ( )
10 2 1

Solution:
5 2(5) − (10) 10 − 10 0 ⃗
a. 𝑇 ( ) = ( )=( )=( )=0
10 −8(5) + 4(10) −40 + 40 0
3 2(3) − (2) 6−2 4
b. 𝑇( ) = ( )=( )=( ) ≠ ⃗0
2 −8(3) + 4(2) −24 + 8 −16
1 2(1) − (1) 2−1 1
c. 𝑇 ( ) = ( )=( ) = ( ) ≠ ⃗0
1 −8(1) + 4(1) −8 + 4 −4

5 3 1
So, ( ) ∈ ker(T) ; ( ) ∉ ker(T) ; ( ) ∉ ker(T)
10 2 1
Theorem 6.6 If 𝑇: 𝑉 → 𝑊 is a linear transformation, then:
1) The kernel of 𝑇 is a subspace of 𝑉.
2) The range of 𝑇 is a subspace of 𝑊.
Proof
a. To prove that 𝐾𝑒(𝑇) is a subspace, we must it is shown that 𝐾𝑒𝑟(𝑇) contains at least one
vector, and is closed to scalar addition and multiplication operations.
According to Definition 6.5, at least vector 0 must be inside 𝐾𝑒(𝑇). Suppose the vectors 𝑣1 and 𝑣2
are inside vectors 𝐾𝑒(𝑇), and taking any scalar 𝛼𝜖ℝ, then:
(𝑣1 + 𝑣2) = 𝑇(𝑣1) + 𝑇(𝑣2) = 0 + 0 = 0,
so that 𝑣1 + 𝑣2 𝜖𝐾𝑒𝑟(𝑇) , and
(𝛼𝑣1) = 𝛼𝑇(𝑣1) = 𝛼. 0 = 0, so 𝛼𝑣1 𝜖 𝐾𝑒(𝑇) .
So it is proven that 𝐾𝑒(𝑇) is a subspace.
b. Since (0) = 0, there is at least one vector in 𝑅𝑎𝑛𝑔𝑒(𝑇).
Let 𝑤1, 𝑤2 be any vectors in 𝑅𝑎𝑛𝑔(𝑇), and take any scalar 𝛼𝜖ℝ. It will be shown that 𝑅𝑎𝑛𝑔(𝑇) is
closed to the scalar addition and multiplication operations. From Definition 6.5, this means that we
must find vectors 𝑣 and 𝑢 in 𝑉, such that (𝑣) = 𝑤1 + 𝑤2 and 𝑇(𝑢) = 𝛼𝑤1.
Because 𝑤1, 𝑤2 𝜖 𝑅𝑎𝑛𝑔(𝑇), it means that there are vectors 𝑣1 𝜖 𝑉 and 𝑣2 𝜖 𝑉 such that 𝑇(𝑣1) = 𝑤1 and
𝑇(𝑣2) = 𝑤2. If 𝑣 = 𝑣1 + 𝑣2 and 𝑢 = 𝛼𝑣1, then:
(𝑣) = 𝑇(𝑣1 + 𝑣2) = 𝑇(𝑣1) + 𝑇(𝑣2) = 𝑤1 + 𝑤2 𝜖 𝑅𝑎𝑛𝑔𝑒(𝑇)
and
(𝑢) = 𝑇( 𝛼𝑣1) = 𝛼 𝑇(𝑣1) = 𝛼𝑤1 𝜖 𝑅𝑎𝑛𝑔𝑒(𝑇) .
So it is proven that 𝑅𝑎𝑛𝑔𝑒 of 𝑇 is a subspace of 𝑊.
Definition 6.7 If 𝑇: 𝑉 → 𝑊 is a linear transformation, then the dimension of 𝑅𝑎𝑛𝑔(𝑇) is called the
rank of 𝑇 and the dimension of 𝐾𝑒𝑟(𝑇) is called the nullity of 𝑇.
Example :
It is known that the linear transformation 𝑇: 𝑅3 → 𝑃2 with

Determine the basis and dimensions of ker(𝑇) and 𝑅(𝑇) (R(T) is the range of T
Solution :

It can get that:

So,

So the transformation matrix for 𝑇 is

By performing an OBE on the enlarged matrix, it is obtained

So (0,0,0) is the only member of ker(T) .So, the basis of ker 𝑇 = { } and the nullity is zero
Then the column space basis of matrix 𝐴 is :

Therefore, the range basis of 𝑇 is:

so that rank (base dimension 𝑅( 𝑡 )=3

B. Linear Tranformation Matrices

In this section, we first discuss coordinate vectors and coordinate matrices. Let 𝐵 = {𝑢
⃗ 1, 𝑢 ⃗ 𝑛 } be
⃗ 2, … , 𝑢
a basis for the vector space V. Since B is a basis in V, it is clear that B is linearly independent. In
section 5.3.2, for any vector 𝑣 in V can be expressed as a linear combination
𝑣 = 𝑘1 𝑢
⃗ 1 + 𝑘2 𝑢
⃗ 2 + ⋯ + 𝑘𝑛 𝑢
⃗𝑛
The scalar 𝑘1 , 𝑘2 , … , 𝑘𝑛 is called the coordinate 𝑣 with respect to base B. This collection of scalars
forms a vector in 𝑅𝑛 which is called the coordinate vector 𝑣 with respect to base B, expressed by (𝑣)𝐵
or can be written as
(𝑣 )𝐵 = (𝑘1 , 𝑘2 , … , 𝑘𝑛 )
Form (6.1) can be written in matrix form, namely
𝑘1
[𝑣]𝐵 = [𝑘…2 ]
𝑘𝑛
where [𝑣 ]𝐵 is the coordinate matrix 𝑣 with respect to base B.

Example 1 Let 𝐵 = {(1,1,1), (1,1,0), (1,0,0)} be a basis for 𝑅3 and a vector 𝑣 = (1,2,1) in 𝑅 3 . We
will find the coordinate vector and coordinate matrix 𝑣 with respect to base B.
For example, 𝑣 can be written as a linear combination.
(1,2,1) = 𝑘1 (1,1,1) + 𝑘2 (1,1,0) + 𝑘3 (1,0,0)
From this linear combination form, three equations are obtained, namely
𝑘1 +𝑘2 + 𝑘3 = 1
𝑘1 +𝑘2 =2
𝑘1 =1
The solution of this system of equations is 𝑘1 = 1., 𝑘2 = 1, 𝑘3 = −1
Therefore, the coordinate vector and coordinate matrix are

1
(𝑣)𝐵 = (1,1, −1) 𝑎𝑛𝑑 [𝑣]𝐵 = [ 1 ]
−1

Theorem 6.8. Let V and W be finite herdimensional vector spaces with respective bases namely B
and B', where 𝐵 = {𝑢 ⃗ 1, 𝑢 ⃗ 𝑛 } and 𝐵 ′ = {𝑣1, 𝑣2 , … , 𝑣𝑚 }. If T:V→W is a linear transformation,
⃗ 2, … , 𝑢
then the transformation representation matrix T is
𝐴 = [[𝑇(𝑢
⃗ 1 )]𝐵′ [𝑇(𝑢
⃗ 2 )]𝐵′ … [𝑇(𝑢
⃗ 𝑛 )]𝐵′ ]
which satisfies the equation [𝑇(𝑢
⃗ 1 )]𝐵′ = 𝐴[𝑣]𝐵 for every 𝑢
⃗ in V.

Let 𝐵 = {𝑢 ⃗ 1, 𝑢 ⃗ 𝑛 } and 𝐵 ′ = {𝑣1 , 𝑣2 , … , 𝑣𝑚 }. The direct method for finding matrix A in Theorem
⃗ 2, … , 𝑢
6.8 is as follows.
(1) Find the matrix [𝑇(𝑢
⃗ 𝑖 )] with 𝑢
⃗ 𝑖 ϵ B.
(2) Construct [𝑇(𝑢
⃗ 𝑖 )] as a linear combination of 𝑣1 , 𝑣2 , … , 𝑣𝑚
[𝑇(𝑢
⃗ 𝑖 )] = 𝑘1 𝑣1 + 𝑘2 𝑣2 + ⋯ + 𝑘𝑚 𝑣𝑚
(3) Complete step (2) until the values of are obtained 𝑘1 , 𝑘2 , … , 𝑘𝑚 .
(4) Create a matrix [𝑇(𝑢
⃗ 1 )]𝐵′ , corresponding to the values of 𝑘1 , 𝑘2 , … , 𝑘𝑚 for respectively 𝑢
⃗ 𝑖 ϵ B i.e
𝑘1
[𝑇(𝑢
⃗ 1 )]𝐵′ = [ 𝑘…2 ]
𝑘𝑚
(5) Create a matrix A whose columns are [𝑇(𝑢
⃗ 1 )]𝐵′ the matrix namely
𝐴 = [[𝑇(𝑢
⃗ 1 )]𝐵′ [𝑇(𝑢
⃗ 2 )]𝐵′ … [𝑇(𝑢
⃗ 𝑛 )]𝐵′ ]
From Theorem 6.8, if 𝐵 ′ = 𝐵 then the representation matrix T is
𝐴 = [[𝑇(𝑢
⃗ 1 )]𝐵′ [𝑇(𝑢
⃗ 2 )]𝐵′ … [𝑇(𝑢
⃗ 𝑛 )] 𝐵 ′ ]
The following example shows that B'=B.

Example 1 For example 𝑇: 𝑅2 → 𝑅2 is a defined linear operator

𝑥1 3𝑥 −𝑥
𝑇 ([𝑥 ]) = [ 1 2 ]
2 𝑥1 + 3𝑥2

Find the representation matrix T with respect to base 𝐵 = {𝑢 ⃗ 2 } and 𝑢

⃗ 1, 𝑢 ⃗ 1 = (1,1) and 𝑢
⃗ 2 = (1, −1)
First find 𝑇(𝑢
⃗ 1 ) and 𝑇(𝑢
⃗ 2 ) which is
1 2 1 4
⃗ 1 ) = 𝑇 [ ] = [ ] 𝑎𝑛𝑑 𝑇(𝑢
𝑇(𝑢 ⃗ 2) = 𝑇 [ ] = [ ]
1 4 −1 −2
⃗ 1 ) as a linear combination of 𝑢
By writing 𝑇(𝑢 ⃗ 1, 𝑢
⃗ 2 i.e
2 1 1
[ ] = 𝑘1 [ ] + 𝑘2 [ ]
4 1 −1
We get 𝑘1 = 3 and 𝑘2 = −1. So that
2
⃗ 1 ) = [ ] = 3𝑢
𝑇(𝑢 ⃗1−𝑢
⃗2
4
Then, by writing as a linear combination of which is
4 1 1
[ ] = 𝑘1 [ ] + 𝑘2 [ ]
−2 1 −1
We get 𝑘1 = 1 and 𝑘2 = 3. So that
4
⃗ 2) = [
𝑇(𝑢 ]=𝑢
⃗ 1 + 3𝑢
⃗2
−2
Therefore
3 1
[𝑇(𝑢
⃗ 1 )]𝐵′ = [ ] , [𝑇(𝑢
⃗ 2 )]𝐵′ = [ ]
−1 3
And
3 1
𝐴 = [[𝑇(𝑢
⃗ 1 )]𝐵′ [𝑇(𝑢
⃗ 2 )]𝐵′ ] = [ ]
−1 3