Paper No.

- 108
Paper Title: Linear Algebra
Definition of Linear transformation
Example 1
Let 𝑇: ℝ2 → ℝ2 be given by 𝑇 𝑥, 𝑦 = 𝑥 + 𝑦, 𝑥 − 𝑦 , ∀ 𝑥, 𝑦 ∈ ℝ2 .
Then show that 𝑇 is a linear map.
Solution:
Example 2
Let 𝑇: ℝ2 → ℝ3 be given by 𝑇 𝑥, 𝑦 = 𝑥, 𝑦, 𝑦 + 2 , ∀ 𝑥, 𝑦 ∈ ℝ2 .
Is 𝑇 a linear map?
Solution:
Example 3
Let 𝑇: ℝ2 → ℝ2 be given by 𝑇 𝑥, 𝑦 = 𝑥𝑦, 𝑦 , ∀ 𝑥, 𝑦 ∈ ℝ2 .
Is 𝑇 a linear map?
Solution:
Theorem 1
Let 𝑉 and 𝑊 be two vector spaces. and 𝑇: 𝑉 → 𝑊 be a linear
transformation. Then

I. 𝑇 𝜃𝑉 = 𝜃𝑊 , where 𝜃𝑉 and 𝜃𝑊 denotes zero vectors in 𝑉 and 𝑊

respectively.
II. 𝑇 −𝑣1 = −𝑇 𝑣1 for 𝑣1 ∈ 𝑉
III. 𝑇 𝑣1 − 𝑣2 = 𝑇 𝑣1 − 𝑇(𝑣2 ) for 𝑣1 , 𝑣2 ∈ 𝑉.
Proof:

𝐼.
𝐼𝐼.

𝐼𝐼𝐼.
Matrix representation of a linear transformation

Defination:
Example 4
Consider a linear map 𝑇: ℝ2 → ℝ3 given by 𝑇 𝑥, 𝑦 = 𝑥, 𝑦, 𝑥 − 𝑦 ,
∀ 𝑥, 𝑦 ∈ ℝ2 . Find a matrix associated to 𝑇 with respect to the standard
basis of ℝ2 and ℝ3 .
Solution:
Example 5
Find a matrix associated to a linear transformation 𝑇: ℝ4 → ℝ2 given by
𝑇 𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 = 𝑥1 + 𝑥3 , 𝑥2 + 𝑥4 , ∀ 𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 ∈ ℝ4 . with respect
to the standard basis of ℝ4 and ℝ2 .
Solution:
Example 6
Let 𝑉 = ℝ2 = 𝑊. Consider a linear map 𝑇: 𝑉 → 𝑊 given by
𝑇 𝑥, 𝑦 = 𝑥 + 𝑦, 𝑥 − 𝑦 , ∀ 𝑥, 𝑦 ∈ 𝑉. Find a matrix associated to 𝑇 with
respect to the basis of 𝐵 = { 0, 2 , 6, 0 } of 𝑉 and 𝐵′ = { 1, 0 , 0, 2 } of
𝑊 respectively.
Solution:
 Paper No.- 108
Paper Title: Linear Algebra
Kernel and image of a linear transformation

Definition:

𝑘𝑒𝑟𝑇 = 𝑣 ∈ 𝑉 ∶ 𝑇 𝑣 = 𝜃𝑊 .

𝐼𝑚𝑇 = 𝑤 ∈ 𝑊 ∶ ∃𝑣 ∈ 𝑉 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑇 𝑣 = 𝑤 .

Note: range of 𝑇,
Theorem 2
Let 𝑉 and 𝑊 be two vector spaces and 𝑇: 𝑉 → 𝑊 be a linear map. Then
I. 𝑘𝑒𝑟𝑇 is a subspace of 𝑉 and
II. 𝐼𝑚𝑇 is a subspace of 𝑊.
Proof:
𝐼.
𝑘𝑒𝑟𝑇 = 𝑣 ∈ 𝑉 ∶ 𝑇 𝑣 = 𝜃𝑊 .

𝑇 is linear map
𝑇 𝛼𝑣1 + 𝑣2 = 𝑇 𝛼𝑣1 + 𝑇 𝑣2 = 𝛼𝑇 𝑣1 + 𝑇(𝑣2 )
𝐼𝐼
𝐼𝑚𝑇 = 𝑤 ∈ 𝑊 ∶ ∃𝑣 ∈ 𝑉 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑇 𝑣 = 𝑤 .

𝑇 is linear map
𝑇 𝛼𝑣1 + 𝑣2 = 𝑇 𝛼𝑣1 + 𝑇 𝑣2 = 𝛼𝑇 𝑣1 + 𝑇(𝑣2 )

Hence 𝐼𝑚𝑇 is a subspace of 𝑊.

Definition
Theorem 3 Rank – Nullity Theorem
Let 𝑉 and 𝑊 be two finite dimensional vector spaces and 𝑇: 𝑉 → 𝑊 be
a linear map. Then
𝑑𝑖𝑚𝑉 = 𝑟𝑎𝑛𝑘 𝑇 + 𝑛𝑢𝑙𝑙𝑖𝑡𝑦 𝑇 .
Proof
𝑘𝑒𝑟𝑇 is a subspace of 𝑉

𝑑𝑖𝑚𝑉 = 𝑘 + 𝑛 = 𝑛𝑢𝑙𝑙𝑖𝑡𝑦 𝑇 + 𝑛.
𝑟𝑎𝑛𝑘(𝑇) = 𝑛.
𝐸 = 𝑇 𝑣1 , 𝑇 𝑣2 , … … … , 𝑇 𝑣𝑛 is a basis for 𝐼𝑚𝑇
𝐸 is linearly independent set
Hence the set 𝐸 is linearly independent set.
show that 𝐼𝑚𝑇 = 𝑆𝑝𝑎𝑛{𝐸}.
∵ 𝑇 is linear map
 ∵ 𝑢1 , 𝑢2 , … , 𝑢𝑘 ∈ 𝑘𝑒𝑟𝑇

∴ 𝐼𝑚𝑇 = 𝑆𝑝𝑎𝑛{𝐸}
Thus, 𝐸 = 𝑇 𝑣1 , 𝑇 𝑣2 , … , 𝑇 𝑣𝑛 is a basis for 𝐼𝑚𝑇.
𝑟𝑎𝑛𝑘 𝑇 = 𝑛.
Hence, 𝑑𝑖𝑚𝑉 = 𝑛𝑢𝑙𝑙𝑖𝑡𝑦 𝑇 + 𝑟𝑎𝑛𝑘(𝑇).
Example 7
Verify Rank – Nullity theorem for a linear map 𝑇: ℝ2 → ℝ2 defined as
𝑇 𝑥, 𝑦 = 𝑥 − 𝑦, 𝑥 + 𝑦 , ∀ 𝑥, 𝑦 ∈ ℝ2 .
Solution:

𝑘𝑒𝑟𝑇 = { 𝑥, 𝑦 ∈ ℝ2 ∶ 𝑇 𝑥, 𝑦 = (0, 0)}

𝐼𝑚𝑇 = { 𝑥1 , 𝑦1 ∈ ℝ2 ∶ ∃ 𝑥, 𝑦 ∈ ℝ2 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑇 𝑥, 𝑦 = (𝑥1 , 𝑦1 )}

Example 8
Verify Rank – Nullity theorem for a linear map 𝑇: ℝ3 → ℝ2 defined as
𝑇 𝑥, 𝑦, 𝑧 = 𝑥 + 𝑦, 𝑦 + 𝑧 , ∀ 𝑥, 𝑦, 𝑧 ∈ ℝ3 .
Solution:

𝑘𝑒𝑟𝑇 = { 𝑥, 𝑦, 𝑧 ∈ ℝ3 ∶ 𝑇 𝑥, 𝑦, 𝑧 = (0, 0)}

𝐼𝑚𝑇 = { 𝑥1 , 𝑦1 ∈ ℝ2 ∶ ∃ 𝑥, 𝑦, 𝑧 ∈ ℝ3 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑇 𝑥, 𝑦, 𝑧 = (𝑥1 , 𝑦1 )}
Example 9
Verify Rank – Nullity theorem for a linear map 𝑇: ℝ3 → ℝ3 defined as
𝑇 𝑥, 𝑦, 𝑧 = 𝑥 + 𝑧, 𝑥 + 𝑦 + 2𝑧, 2𝑥 + 𝑦 + 3𝑧 , ∀ 𝑥, 𝑦, 𝑧 ∈ ℝ3 .
Solution:
 Online Education by

Department of Mathematics
M K Bhavnagar University
For M.Sc. (Mathematics) Sem-2
 Paper No.: 108
Linear Algebra
Instructor: Dr. P. I. Andharia
 Syllabus: Unit-2
• Linear Transformations
• Representation of Linear
Transformations by Matrices
• Rank-Nullity Theorem
• Duality and Transpose
• Determinants
• Co-factors, Adjunct, Cramer’s Rule.
 0 ≠
Definition: A Kronecker delta is defined as =
1 =
Definition: A linear map from the vector space into the scalar field
is called a linear functional on .
Definition: Let be a vector space. The collection of all linear functional
on forms a vector space and it is denoted by ( , ) or ∗ and we call
it the dual space of .
Remarks:
(1) dim ∗ = dim .
(2) Let be a vector space and = { 1 , 2 , … … , } is a basis for .
Then ∃ linear functional : → such that = and the
set { 1 , 2 , … … , } form a basis for ∗ . This basis is called dual
basis of .
Theorem: Let , and be vector spaces over the same field . Let
: → and : → are linear transformations then the
composition map : → is a linear transformation.
Proof: Let 1, 2 ∈ and ∈ .
( )( 1 + 2) = [ ( 1 + 2 )]
= [ ( 1) + ( 2 )] ∵ is L. T.
= [ ( 1 )] + [ ( 2 )] ∵ is L. T.
=( )( 1 ) + ( )( 2)
( )( 1) = [ ( 1 )]
= [ ( 1 )] ∵ is L. T.
= [ ( 1 )] ∵ is L. T.
= ( )( 1 )
So, the composition map is linear transformation.
Theorem:
Let and be vector spaces over the field and : → be a linear
∗
transformation then ∃ a unique linear transformation : → ∗ such
that ( )( ) = ( ( )) where is a linear functional on .
Proof:
∗ ∗
Let and be duals of and respectively.
∗
Suppose is a linear functional on then : → and ∈ .
∗
Consider = then : → and ∈ .
∗ ∗ ( )= ∗
Define : → as = , ∀ ∈ . Then,
( )( ) = ( ) , ∈ . ( is known as transpose of )
To prove: is linear transformation.
∗
Let 1, 2 ∈ , ∈ and ∈ then
 [ ( 1 + 2 )]( ) =( 1 + 2) ( )
= 1( ( )) + 2( ( ))
=( 1 )( )+( 2 )( )
=( 1 + 2 )( )
∴ ( 1 + 2) = ( 1) + ( 2)
[ ( 1 )]( ) =( 1) ( )
= ( 1( ( ) ))
= ( 1 )( )
= ( 1 )( )
= ( 1 )( )
∴ ( 1) = ( 1)
So, is linear transformation. This completes the proof.
Definition:
Let be a scalar field and be a positive integer. Let be a function
defined on × matrices which assigns a scalar in the field . Then
is said to be -linear if ∀ 1 ≤ ≤ , is linear function of ℎ raw when
remaining − 1 raws are fixed. i.e. if we suppose the matrix
= ( 1, 2, … … , ) where 1, 2, … … , ∈ are rows of then

1, 2, … … , + ,……,
= ( 1, 2, … … , ,……, )+ 1, 2, … … , ,……, .
Definition: Let be -linear function. We say is alternating if
(i) ( ) = 0 whenever two rows of are equal i.e.
( 1, 2, … … , , … … , , … … , ) = 0
(ii) If ′ is a matrix obtained from by interchanging two rows of
then ( ′ ) = − ( ) i.e.
1, 2, … … , , … … , , … … ,
=− 1, 2, … … , ,……, ,……, .
Definition: Let be a scalar field and be a positive integer. Let be a
function defined on × matrices which assigns a scalar in the field .
Then is called determinant function if is -linear, alternating and
( ) = 1, where is identity matrix of order × . It is denoted by .
Note: (1) ( 1, 2, … … , ) = 1, where ( 1 , 2, … … , ) is basis for .
(2) ( 1, 2, … … , ) = 0, if ′ are linearly dependent.
Computation of Determinant:
11 12
Example: Find determinant of a matrix = .
21 22
Solution:
det = det( 11 1 + 21 2 , 12 1 + 22 2 )
= det( 11 1 , 12 1 + 22 2 ) + det( 21 2 , 12 1 + 22 2 )

= det( 11 1 , 12 1 ) + det( 11 1 , 22 2 ) + det( 21 2 , 12 1 )

+ det( 21 2 , 22 2 )

= det( 11 1 , 22 2 ) + det( 21 2 , 12 1 )

∵ det( 11 1 , 12 1 ) and det( 21 2 , 22 2 ) are linearly dependent.

= 11 22 det( 1 , 2) + 21 12 det( 2 , 1)

= 11 22 det( 1 , 2) − 21 12 det( 1 , 2)

= 11 22 − 21 12 ∵ det( 1 , 2) =1
 11 0…… 0
0 22…… 0
Example: Find determinant of a matrix = .
: : :
0 0……
Solution:
det = det( 11 1 , 22 2 , … … , )
= 11 22 …… det( 1 , 2 , … … , )
= 11 22 … …

Example: Find determinant of a matrix = .

Solution:
det = det( 1 + 2 + 3 , 1 + 2 + 3 , 1 + 2 + 3 )
= det( 1 , 2 , 3 ) + det( 1 , 3 , 2 ) + det( 2 , 1 , 3 )
+ det( 2 , 3 , 1 ) + det( 3 , 1 , 2 ) + det( 3 , 2 , 1 )
 = det( 1 , 2 , 3 ) + det( 1 , 3 , 2 ) + det( 2 , 1 , 3 )
+ det( 2 , 3 , 1 ) + det( 3 , 1 , 2 ) + det( 3 , 2 , 1 )
= det( 1 , 2 , 3 ) − det( 1 , 2 , 3 ) − det( 1 , 2 , 3 )
+ det( 1 , 2 , 3 ) + det( 1 , 2 , 3 ) − det( 1 , 2 , 3 )
= + + − − −
= ( − )− ( − )+ ( − )

= det − det + det

Definition: Let =[ ] be × matrix. Let denote the square

ℎ ℎ
matrix of order ( − 1) obtained by deleting the row and column
from the matrix . Then
(1) An × matrix =[ ] is called transpose of if = . It is
denoted by .
(2) det is called the minor of the element of .
 (3) The co-factor of the element of is defined as
= (−1) + det
and the matrix [ ] is called the co-factor matrix of .
(4) The transpose of co-factor matrix of is known as adjunct of
and it is denoted by ̅ or .
1
(5) The inverse matrix of is defined as −1 = ̅.
det

Example:
0 −1 3
−1
Let = 2 5 −4 then find .
−3 7 1
Solution: From the given matrix , we get
5 −4
11 = = 5 + 28 = 33, ∴ 11 = (−1)1+1 11 = 33
7 1
 2 −4
12= = 2 − 12 = −10, ∴ 12 = (−1)1+2 12 = 10
−3 1
2 5
13 = = 14 + 15 = 29, ∴ 13 = (−1)1+3 13 = 29
−3 7
−1 3
21 = = −1 − 21 = −22, ∴ 21 = (−1)2+1 21 = 22
7 1
0 3
22 = = 0 + 9 = 9, ∴ 22 = (−1)2+2 22 = 9
−3 1
0 −1
23 = = 0 − 3 = −3, ∴ 23 = (−1)2+3 23 = 3
−3 7
−1 3
31 = = 4 − 15 = −11, ∴ 31 = (−1)3+1 31 = −11
5 −4
0 3
32 = = 0 − 6 = −6, ∴ 32 = (−1)3+2 32 = 6
2 −4
0 −1
33 = = 0 + 2 = 2, ∴ 33 = (−1)3+3 33 = 2
2 5
 33 10 29
So, Co-factor matrix of is 22 9 3 .
−11 6 2
∴ ̅= = Transpose of co-factor matrix of
33 22 −11
= 10 9 6
29 3 2
0 −1 3
Also, we have det = det 2 5 −4
−3 7 1
= 0(33) − (−1)(−10) + 3(29) = 77
33 22 −11
1 1
Hence, −1 = ̅= 10 9 6 .
det 77
29 3 2

−1 1
Note: (1) det( ) = (det ) (det ) (2) det = (det )−1 =
det
Notation: The class of all × matrices with all real entries is denoted
by ( , ).
Theorem: Cramer’s Rule:
Let ∈ ( , ) with det ≠ 0. Let = ( 1 , 2 , … … , ) be a column
vector and a column vector ∈ then the solution of = is given
det ( 1 , 2 ,……, −1 , , +1 ,……, ) ℎ
by = , where is the column of .
det

Proof:
Let = ( 1 , 2, … … , ) be identity matrix in .
We want to solve = , where is the fixed column vector.
Write =( 1, 2, … … , ).
Take = ( 1, 2, … … , −1 , , +1 , … … , ) then
det = .
Now, =( 1, 2, … … , −1 , , +1 , … … , )
=( 1, 2, … … , −1 , , +1 , … … , )
But, = det
= det( )
−1
= det( )
−1 )
= det( det( )
= (det )−1 det( )
1
= det( )
det
1
= det( 1, 2, … … , −1 , , +1 , … … , )
det

This completes the proof.

Example:
Solve the system of equations using Cramer’s rule:
+ =0
+ =1
+ = −1
Solution:
Let us write the given system of equations in matrix form as
=
1 1 0 0
where, = 0 1 1 , = and = 1 .
1 0 1 −1
First we compute the determinant of .
 1 1 0
( ) 1 1 0 1
det = det 0 1 1 = 1 det − (1) det = 2 ≠ 0.
0 1 1 1
1 0 1
Now, we apply the Cramer’s rule. Let denote the matrix obtained
from by replacing the ℎ column by the column vector . Then,
0 1 0
1 =( , 2, 3) = 1 1 1 .
−1 0 1
1 1
Hence, det = −(1) det
1 = −(1 + 1) = −2.
−2
−1 1
1 0 0
Similarly, 2 = ( 1 , , 3 ) = 0 1 1 .
1 −1 1
1 1
Hence, det 2 = (1) det = 1 + 1 = 2.
−1 1
 1 1 0
And 3 = ( 1 , 2 , ) = 0 1 1 .
1 0 −1
1 1 0 1
Hence, det 3 = (1) det − (1) det = −1 + 1 = 0.
0 −1 1 −1
The solution is given by
det 1 −2
= = = −1
det 2
det 2 2
= = =1
det 2
det 3 0
= = = 0.
det 2
Hence, the required solution is {−1, 1, 0}.