LINEAR ALGEBRA (BCS405A)

Module 1
The geometry of linear equations: existence and uniqueness, LU-decomposition and
computational complexity, Vector Spaces and subspaces (Definitions only), The column space
and null space of a given matrix, linearly independent, Bases and dimension, Null space and
column space of a matrix, The four fundamental subspaces.
Textbook 2: Ch - 1.2, 1.5, 2.1, 2.3, 2.4

1.1 Geometry of linear equations – Existence and uniqueness

I. Homogeneous system of linear equations

Consistent Solution exists

Inconsistent Solution does not exist
𝑎1 𝑥 + 𝑏1 𝑦 + 𝑐1 𝑧 = 0
Homogeneous system 𝑎2 𝑥 + 𝑏2 𝑦 + 𝑐2 𝑧 = 0
𝑎3 𝑥 + 𝑏3 𝑦 + 𝑐3 𝑧 = 0
𝑎1 𝑏1 𝑐1 𝑥 0
Matrix form ( 𝑎 2 𝑏 2 𝑐2 ) (𝑦 ) = ( 0)
𝑎3 𝑏3 𝑐3 𝑧 0
or
𝐴𝑋 = 0
Coefficient matrix 𝑎1 𝑏1 𝑐1
𝐴 = 2 𝑏2 𝑐2 )
( 𝑎
𝑎3 𝑏3 𝑐3
𝜌(𝐴) = 3 Trivial solution (Unique)
𝑥 = 0, 𝑦 = 0, 𝑧 = 0
𝜌(𝐴) < 3 Non-trivial solution (Infinite)
𝑧 = 𝑘, 𝑦 = 𝑝𝑘, 𝑥 = 𝑞𝑘

Homogeneous system of equations is always consistent.

Dr. Narasimhan G, RNSIT 1

 II. Non-homogeneous system of linear equations

𝑎1 𝑥 + 𝑏1 𝑦 + 𝑐1 𝑧 = 𝑑1
Non-homogeneous 𝑎2 𝑥 + 𝑏2 𝑦 + 𝑐2 𝑧 = 𝑑2
system 𝑎3 𝑥 + 𝑏3 𝑦 + 𝑐3 𝑧 = 𝑑3
𝑎1 𝑏1 𝑐1 𝑥 𝑑1
(𝑎 2 𝑏2 𝑐2 ) (𝑦) = (𝑑2 )
Matrix form
𝑎3 𝑏3 𝑐3 𝑧 𝑑3
or
𝐴𝑋 = 𝐵
Augmented matrix 𝑎1 𝑏1 𝑐1 𝑑1
(𝐴, 𝐵) = (𝑎2 𝑏2 𝑐2 𝑑2 )
𝑎3 𝑏3 𝑐3 𝑑3
𝜌(𝐴) = 𝜌(𝐴, 𝐵) Consistent
𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 3 Unique solution
𝜌(𝐴) = 𝜌(𝐴, 𝐵) < 3 Infinite number of solutions
𝜌(𝐴) ≠ 𝜌(𝐴, 𝐵) Inconsistent

1. Test for consistency and solve the following system of equations:

𝒙 + 𝟐𝒚 + 𝟑𝒛 = 𝟎, 𝟑𝒙 + 𝟒𝒚 + 𝟒𝒛 = 𝟎, 𝟕𝒙 + 𝟏𝟎𝒚 + 𝟏𝟐𝒛 = 𝟎
Coefficient matrix is
1 2 3
𝐴 = (3 4 4)
7 10 12
𝑅2 → 𝑅2 − 3𝑅1 , 𝑅3 → 𝑅3 − 7𝑅1
1 2 3
~ (0 −2 −5)
0 −4 −9
𝑅3 → 𝑅3 − 2𝑅2
1 2 3
~ (0 −2 −5)
0 0 1
This is in echelon form. Number of non-zero rows is 3.
𝜌(𝐴) = 3. Therefore, the given system of equations has trivial solution.
Therefore, 𝑥 = 0, 𝑦 = 0, 𝑧 = 0.

Dr. Narasimhan G, RNSIT 2

2. Test for consistency and solve the following system of equations:
𝒙 + 𝟑𝒚 − 𝟐𝒛 = 𝟎, 𝟐𝒙 − 𝒚 + 𝟒𝒛 = 𝟎, 𝒙 − 𝟏𝟏𝒚 + 𝟏𝟒𝒛 = 𝟎
Coefficient matrix is
1 3 −2
𝐴 = (2 −1 4)
1 −11 14
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 𝑅1
1 3 −2
~ (0 −7 8)
0 −14 16
𝑅3 → 𝑅3 − 2𝑅2
1 3 −2
~ (0 −7 8)
0 0 0
This is in echelon form. Number of non-zero rows is 2.
𝜌(𝐴) = 2. Therefore, the given system of equations has non-trivial solutions.
Reduced system of equations is
𝑥 + 3𝑦 − 2𝑧 = 0 ---- (1)7𝑦 + 8𝑧 = 0 ------- (2)
8𝑘 10
Choose 𝑧 = 𝑘 then 𝑦 = and 𝑥 = − 𝑘
7 7
10 8𝑘
Therefore, 𝑥 = − 𝑘 ,𝑦 = ,𝑧 = 𝑘
7 7

Dr. Narasimhan G, RNSIT 3

3. Test for consistency and solve the following system of equations:
𝒙 + 𝟑𝒚 + 𝟐𝒛 = 𝟎, 𝟐𝒙 − 𝒚 + 𝟑𝒛 = 𝟎, 𝟑𝒙 − 𝟓𝒚 + 𝟒𝒛 = 𝟎, 𝒙 + 𝟏𝟕𝒚 + 𝟒𝒛 = 𝟎

Coefficient matrix is
1 3 2
𝐴 = (2 −1 3)
3 −5 4
1 17 4
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 3𝑅1 , 𝑅4 → 𝑅4 − 𝑅1
1 3 2
~(0 −7 −1 )
0 −14 −2
0 14 2
𝑅3 → 𝑅3 − 2𝑅2
1 3 2
~ (0 −7 −1)
0 0 0
0 0 0
This is in echelon form. Number of non-zero rows is 2.
𝜌(𝐴) = 2. Therefore, the given system of equations has non-trivial solutions.

Reduced system of equations is

𝑥 + 3𝑦 + 2𝑧 = 0 ---- (1)
−7𝑦 − 𝑧 = 0 ----- (2)
−𝑘 11
Choose 𝑧 = 𝑘 then 𝑦 = and 𝑥 = − 𝑘
7 7
11 −𝑘
Therefore, 𝑥 = − 𝑘 ,𝑦 = ,𝑧 = 𝑘
7 7

Dr. Narasimhan G, RNSIT 4

4. Test for consistency and solve the following system of equations:
𝒙 + 𝟐𝒚 + 𝟑𝒛 = 𝟏, 𝟐𝒙 + 𝟑𝒚 + 𝟖𝒛 = 𝟐, 𝒙 + 𝒚 + 𝒛 = 𝟑

Coefficient matrix is
1 2 3 1
(𝐴, 𝐵) = (2 3 8 2)
1 1 1 3
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 𝑅1
1 2 3 1
~ (0 −1 2 0)
0 −1 −2 2
𝑅3 → 𝑅3 − 𝑅2
1 2 3 1
~ (0 −1 2 0)
0 0 −4 2
This is in echelon form. Number of non-zero rows is 3.
𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 3.
Therefore, the given system of equations is consistent and has a unique solution.
Reduced system of equations is
𝑥 + 2𝑦 + 3𝑧 = 1 ---- (1)
−𝑦 + 2𝑧 = 0 ---- (2)
−4𝑧 = 2 ---- (3)
Solving by back substitution,
9 1
𝑥 = 2 , 𝑦 = −1, 𝑧 = − 2

Dr. Narasimhan G, RNSIT 5

5. Test for consistency and solve the following system of equations:
𝟓𝒙 + 𝟑𝒚 + 𝟕𝒛 = 𝟒, 𝟑𝒙 + 𝟐𝟔𝒚 + 𝟐𝒛 = 𝟗, 𝟕𝒙 + 𝟐𝒚 + 𝟏𝟎𝒛 = 𝟓

Coefficient matrix is
5 3 7 4
(𝐴, 𝐵) = (3 26 2 9)
7 2 10 5
𝑅2 → 5𝑅2 − 3𝑅1 , 𝑅3 → 5𝑅3 − 7𝑅1
5 3 7 4
~ (0 121 −11 33 )
0 −11 1 −3
𝑅3 → 11𝑅3 + 𝑅2
5 3 7 4
~ (0 121 −11 33)
0 0 0 0
This is in echelon form. Number of non-zero rows is 2.
𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 2.
Therefore, the given system of equations is consistent
and has an infinite number of solutions.
Reduced system of equations is
5𝑥 + 3𝑦 + 7𝑧 = 4 ---- (1)
121𝑦 − 11𝑧 = 33 ---- (2)
3+𝑘 7−16𝑘
Choose 𝑧 = 𝑘 then 𝑦 = and 𝑥 = .
11 11
7−16𝑘 3+𝑘
Therefore, 𝑥 = ,𝑦 = , 𝑧 = 𝑘.
11 11

Dr. Narasimhan G, RNSIT 6

6. Test for consistency and solve the following system of equations:
𝒙 + 𝒚 + 𝒛 = −𝟑, 𝟑𝒙 + 𝒚 − 𝟐𝒛 = −𝟐, 𝟐𝒙 + 𝟒𝒚 + 𝟕𝒛 = 𝟕
Coefficient matrix is
1 1 1 −3
(𝐴, 𝐵) = (3 1 −2 −2)
2 4 7 7
𝑅2 → 𝑅2 − 3𝑅1 , 𝑅3 → 𝑅3 − 2𝑅1
1 1 1 −3
~ (0 −2 −5 7 )
0 2 5 13
𝑅3 → 𝑅3 + 𝑅2
1 1 1 −3
~ (0 −2 −5 7 )
0 0 0 20
This is in echelon form. Number of non-zero rows is 3.
𝜌(𝐴) = 2 𝑎𝑛𝑑 𝜌(𝐴, 𝐵) = 3. 𝜌(𝐴) ≠ 𝜌(𝐴, 𝐵).
Therefore, the given system of equations is inconsistent
Therefore, the system has no solution.

7. Test for consistency and solve the following system of equations:

𝒙 + 𝒚 + 𝒛 = 𝟗, 𝒙 − 𝟐𝒚 + 𝟑𝒛 = 𝟖, 𝟐𝒙 + 𝒚 − 𝒛 = 𝟑
Coefficient matrix is
1 1 1 9
(𝐴, 𝐵) = (1 −2 3 8)
2 1 −1 3
𝑅2 → 𝑅2 − 𝑅1 , 𝑅3 → 𝑅3 − 2𝑅1
1 1 1 9
~ (0 −3 2 −1 )
0 −1 −3 −15
𝑅3 → 3𝑅3 − 𝑅2
1 1 1 9
~ (0 −3 2 −1 )
0 0 −11 −44
This is in echelon form. Number of non-zero rows is 3.
𝜌(𝐴) = 3 𝑎𝑛𝑑 𝜌(𝐴, 𝐵) = 3. 𝜌(𝐴) = 𝜌(𝐴, 𝐵).
Therefore, the given system of equations is consistent

Dr. Narasimhan G, RNSIT 7

 Therefore, the system has a unique solution.

8. Test for consistency and solve the following system of equations:

𝒙 + 𝟐𝒚 + 𝟐𝒛 = 𝟓, 𝟐𝒙 + 𝒚 + 𝟑𝒛 = 𝟔, 𝟑𝒙 − 𝒚 + 𝟐𝒛 = 𝟒, 𝒙 + 𝒚 + 𝒛 = −𝟏

Coefficient matrix is
1 2 2 5
(𝐴, 𝐵) = ( 2 1 3 6)
3 −1 2 4
1 1 1 −1
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 3𝑅1 , 𝑅4 → 𝑅4 − 𝑅1
1 2 2 5
~ (0 −3 −1 −4 )
0 −7 −4 −11
0 −1 −1 −6
𝑅3 → 3𝑅3 − 7𝑅2 , 𝑅4 → 3𝑅4 − 𝑅2
1 2 2 5
~( 0 −3 −1 −4 )
0 0 −5 −5
0 0 −2 − 14
𝑅4 → 5𝑅4 − 2𝑅3
1 2 2 5
~( 0 −3 −1 −4 )
0 0 −5 −5
0 0 0 − 60
This is in echelon form. Number of non-zero rows is 4.
𝜌(𝐴) = 3 𝑎𝑛𝑑 𝜌(𝐴, 𝐵) = 4. 𝜌(𝐴) ≠ 𝜌(𝐴, 𝐵).
Therefore, the given system of equations is inconsistent
Therefore, the system has no solution.

Dr. Narasimhan G, RNSIT 8

9. Find the values of 𝝀 and 𝝁 for which the system
𝒙 + 𝒚 + 𝒛 = 𝟔, 𝒙 + 𝟐𝒚 + 𝟑𝒛 = 𝟏𝟎, 𝒙 + 𝟐𝒚 + 𝝀𝒛 = 𝝁
has (i) Unique solution (ii) Infinitely many solutions (iii) No solution.
(May 22)
Coefficient matrix is
1 1 1 6
(𝐴, 𝐵) = (1 2 3 10)
1 2 𝜆 𝜇
𝑅2 → 𝑅2 − 𝑅1 , 𝑅3 → 𝑅3 − 𝑅1
1 1 1 6
~ (0 1 2 4 )
0 1 𝜆−1 𝜇−6
𝑅3 → 𝑅3 − 𝑅2
1 1 1 −3
~ (0 −2 −5 7 )
0 0 𝜆 − 3 𝜇 − 10
This is in echelon form.
(i) If 𝜆 ≠ 3, then 𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 3.
The given system of equations has a unique solution.
(ii) If 𝜆 = 3, 𝜇 = 10 then 𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 2.
The given system of equations has infinitely many solutions.
(iii) If 𝜆 = 3, 𝜇 ≠ 10 then 𝜌(𝐴) ≠ 𝜌(𝐴, 𝐵).
The given system of equations has no solution.

Dr. Narasimhan G, RNSIT 9

 10. Find the values of 𝝀 and 𝝁 for which the system
𝟐𝒙 + 𝟑𝒚 + 𝟓𝒛 = 𝟗, 𝟕𝒙 + 𝟑𝒚 − 𝟐𝒛 = 𝟖, 𝟐𝒙 + 𝟑𝒚 + 𝝀𝒛 = 𝝁
has (i) Unique solution (ii) Infinitely many solutions (iii) No solution.

Coefficient matrix is
2 3 5 9
(𝐴, 𝐵) = (7 3 −2 8)
2 3 𝜆 𝜇
𝑅2 → 2𝑅2 − 7𝑅1 , 𝑅3 → 𝑅3 − 𝑅1
1 1 1 6
~ (0 −15 −39 −47 )
0 0 𝜆−5 𝜇−9

This is in echelon form.

(i) If 𝜆 ≠ 5, then 𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 3.
The given system of equations has a unique solution.
(ii) If 𝜆 = 5, 𝜇 = 9 then 𝜌(𝐴) = 𝜌(𝐴, 𝐵) = 2.
The given system of equations has infinitely many solutions.
(iii) If 𝜆 = 5, 𝜇 ≠ 9 then 𝜌(𝐴) ≠ 𝜌(𝐴, 𝐵).
The given system of equations has no solution.

Dr. Narasimhan G, RNSIT 10

 1.2 LU-decomposition and computational complexity

Introduction:

❖ LU decomposition is an important matrix factorization technique in the Numerical Linear

Algebra.
❖ It expresses a matrix 𝐴 as a product of the lower and upper triangular matrices.
𝐴 = 𝐿𝑈, where 𝐿 is a lower triangular matrix and 𝑈is an upper triangular matrix.
❖ It helps to compute the determinant of the given matrix easily. 𝐷𝑒𝑡 𝐴 is the product of
the diagonal elements of U.
❖ LU decomposition turns a large matrix problem into simple triangular systems, making
computations faster, reusable, and more reliable.

Note: LU decomposition of a matrix is not possible (without pivoting) when:

❖ A pivot element becomes zero during elimination.

❖ Any leading principal minor is zero. (Determinant of the left 𝑘 × 𝑘 submatrix)
❖ The matrix is singular (det 𝐴 = 0).
❖ Row interchanges are required (then only 𝑃𝐴 = 𝐿𝑈exists).
❖ The matrix is ill-conditioned (|𝐴| → 0)with very small pivots.

𝟐 𝟓
1. Find an LU decomposition of ( )
−𝟑 −𝟒
2 5
Let 𝐴 = ( )
−3 −4
1
First column of 𝐿 = 2 (First column of A), 2 is the pivot.

First row of U = First row of A

𝐴 = 𝐿𝑈
2 5 1 0 2 5
( ) = (− 3 1) ( )
−3 −4 2 0 𝑥

2 5 2 5
( ) = (−3 𝑥 − 15)
−3 −4 2
15
Equating components, −4 = 𝑥 − 2

Dr. Narasimhan G, RNSIT 11

 7
Therefore, 𝑥 = 2

2 5 1 0 2 5
Therefore, LU decomposition A is ( ) = (− 3 1) (0 7)
−3 −4 2 2
𝟔 𝟒
2. Find the LU factorisation of ( )
𝟏𝟐 𝟓

6 4
Let 𝐴 = ( )
12 5
1
First column of 𝐿 = 6 (First column of A), 6 is the pivot.

First row of U = First row of A

𝐴 = 𝐿𝑈
6 4 1 0 6 4
( )=( )( )
12 5 2 1 0 𝑥
6 4 6 4
( )=( )
12 5 12 8+𝑥
Equating components, 5 = 8 + 𝑥
𝑥 = −3
Therefore, LU decomposition A is
6 4 1 0 6 4
( )=( )( )
12 5 2 1 0 −3
3 1 2
3. Find an LU decomposition of (−9 0 −4)
9 9 14

3 1 2
Let 𝐴 = (−9 0 −4)
9 9 14
1
First column of 𝐿 = 3 (First column of A), 3 is the pivot.

First row of U = First row of A

𝐴 = 𝐿𝑈
3 1 2 1 0 0 3 1 2
(−9 0 −4) = (−3 1 0 ) (0 𝑥 𝑦 )
9 9 14 3 𝑎 1 0 0 𝑧
3 1 2 3 1 2
(−9 0 −4) = (−9 −3 + 𝑥 −6 + 𝑦 )
9 9 14 9 3 + 𝑎𝑥 6 + 𝑎𝑦 + 𝑧
Equating components, 0 = −3 + 𝑥, −4 = −6 + 𝑦, 9 = 3 + 𝑎𝑥
Therefore, 𝑥 = 3, 𝑦 = 2, 𝑎 = 2, 𝑧 = 4
Therefore, LU decomposition A is

Dr. Narasimhan G, RNSIT 12

 3 1 2 1 0 0 3 1 2
(−9 0 −4 ) = ( −3 1 0) (0 3 2)
9 9 14 3 2 1 0 0 4
−5 0 4
4. Find the LU factorisation of ( 10 2 −5)
10 10 16

−5 0 4
Let 𝐴 = ( 10 2 −5)
10 10 16
1
First column of 𝐿 = −5 (First column of A), -5 is the pivot.

First row of U = First row of A

𝐴 = 𝐿𝑈
−5 0 4 1 0 0 −5 0 4
( 10 2 −5 ) = (−2 1 0) ( 0 𝑥 𝑦)
10 10 16 −2 𝑎 1 0 0 𝑧
−5 0 4 −5 0 4
( 10 2 −5) = ( 10 𝑥 −8 + 𝑦 )
10 10 16 10 𝑎𝑥 −8 + 𝑎𝑦 + 𝑧
Equating components, 2 = 𝑥, −5 = −8 + 𝑦, 10 = 𝑎𝑥, 16 = −8 + 𝑎𝑦 + 𝑧
Therefore, 𝑥 = 2, 𝑦 = 3, 𝑎 = 5, 𝑧 = 9
Therefore, LU decomposition A is
−5 0 4 1 0 0 −5 0 4
( 10 2 −5) = (−2 1 0) ( 0 2 3)
10 10 16 −2 5 1 0 0 9

Dr. Narasimhan G, RNSIT 13

 3 7 2
5. Find an LU decomposition of ( 6 19 4)
−2 −2 3

3 7 2
Let 𝐴 = ( 6 19 4)
−2 −2 3
1
First column of 𝐿 = 3 (First column of A), 3 is the pivot.

First row of U = First row of A

𝐴 = 𝐿𝑈
3 7 2 1 0 0 3 7 2
( 6 19 4) = ( 22 1 0) (0 𝑥 𝑦)
−2 −2 3 −3 𝑎 1 0 0 𝑧

3 7 2 3 7 2
( 6 19 4) =( 6 14 + 𝑥 4+𝑦 )
14 4
−2 −2 3 −2 − 3 + 𝑎𝑥 − 3 + 𝑎𝑦 + 𝑧
14 4
Equating components, 19 = 14 + 𝑥, 4 = 4 + 𝑦, −2 = − + 𝑎𝑥, 3 = − 3 + 𝑎𝑦 + 𝑧
3
8 13
Therefore, 𝑥 = 5, 𝑦 = 0, 𝑎 = 15 , 𝑧 = 3

Therefore, LU decomposition A is
3 7 2 1 0 0 3 7 2
( 6 19 4) = ( 22 1
8
0) ( 0 5 0)
13
−2 −2 3 −3 1 0 0
15 3

Dr. Narasimhan G, RNSIT 14

 𝟑 −𝟕 −𝟐 −𝟕
6. If 𝑨 = (−𝟑 𝟓 𝟏 ) ,𝒃 = ( 𝟓 ) ,
𝟔 −𝟒 𝟎 𝟐
solve the equation 𝑨𝒙 = 𝒃 by using the LU factorization.

3 −7 −2 1 0 0 3 −7 −2
(−3 5 1 ) = (−1 1 0) (0 𝑥 𝑦)
6 −4 0 2 𝑎 1 0 0 𝑧

3 −7 −2 3 −7 −2
(−3 5 1 ) = (−3 7+𝑥 2+𝑦 )
6 −4 0 6 −14 + 𝑎𝑥 −4 + 𝑎𝑦 + 𝑧

Equating components, 𝑥 = −2, 𝑦 = −1, 𝑎𝑥 = 10, 𝑧 = 4 − 𝑎𝑦

Therefore, 𝑥 = −2, 𝑦 = −1, 𝑎 = −5, 𝑧 = −1

1 0 0 3 −7 −2
Therefore, 𝑳 = (−1 1 0) and 𝑈 = (0 −2 −1)
−2 −5 1 0 0 −1

𝐴𝑥 = 𝑏
(𝐿𝑈)𝑥 = 𝑏
𝐿(𝑈𝑥) = 𝑏
𝐿𝑦 = 𝑏, where 𝑦 = 𝑈𝑥
1 0 0 𝑦1 −7
(−1 1 0) (𝑦2 ) = ( 5 )
2 −5 1 𝑦3 2
𝑦1 = −7
−𝑦1 + 𝑦2 = 5
2𝑦1 − 5𝑦2 + 𝑦3 = 2
By solving, 𝑦1 = −7, 𝑦2 = −2, 𝑦3 = 6
𝑈𝑥 = 𝑦
3 −7 −2 𝑥1 −7
𝑥
(0 −2 −1) ( 2 ) = (−2)
0 0 −1 𝑥3 6
3𝑥1 − 7𝑥2 − 2𝑥3 = −7
−2𝑥2 − 𝑥3 = −2
−𝑥3 = 6
By solving, 𝑥1 = 3, 𝑥2 = 4, 𝑥3 = −6
3
The solution is 𝑥 = ( 4 )
−6

Dr. Narasimhan G, RNSIT 15

 𝟐 −𝟔 𝟒 𝟐
7. If 𝑨 = (−𝟒 𝟖 𝟎) , 𝒃 = (−𝟒),
𝟎 −𝟒 𝟔 𝟔
solve the equation 𝑨𝒙 = 𝒃 by using the LU factorization.

2 −6 4 1 0 0 2 −6 4
(−4 8 0) = (−2 1 0) (0 𝑥 𝑦)
0 −4 6 0 𝑎 1 0 0 𝑧

2 −6 4 2 −6 4
(−4 8 0) = (−3 7+𝑥 2+𝑦 )
0 −4 6 6 −14 + 𝑎𝑥 −4 + 𝑎𝑦 + 𝑧

Equating components, 𝑥 = −2, 𝑦 = −1, 𝑎𝑥 = 10, 𝑧 = 4 − 𝑎𝑦

Therefore, 𝑥 = −2, 𝑦 = −1, 𝑎 = −5, 𝑧 = −1

1 0 0 3 −7 −2
Therefore, 𝑳 = (−1 1 0) and 𝑈 = (0 −2 −1)
−2 −5 1 0 0 −1

𝐴𝑥 = 𝑏
(𝐿𝑈)𝑥 = 𝑏
𝐿(𝑈𝑥) = 𝑏
𝐿𝑦 = 𝑏, where 𝑦 = 𝑈𝑥
1 0 0 𝑦1 −7
(−1 1 0) (𝑦2 ) = ( 5 )
2 −5 1 𝑦3 2
𝑦1 = −7
−𝑦1 + 𝑦2 = 5
2𝑦1 − 5𝑦2 + 𝑦3 = 2
By solving, 𝑦1 = −7, 𝑦2 = −2, 𝑦3 = 6
𝑈𝑥 = 𝑦
3 −7 −2 𝑥1 −7
(0 𝑥
−2 −1) ( 2 ) = (−2)
0 0 −1 𝑥3 6
3𝑥1 − 7𝑥2 − 2𝑥3 = −7
−2𝑥2 − 𝑥3 = −2
−𝑥3 = 6
By solving, 𝑥1 = 3, 𝑥2 = 4, 𝑥3 = −6
3
The solution is 𝑥 = ( 4 )
−6

Dr. Narasimhan G, RNSIT 16

 𝟐 −𝟒 𝟐 𝟔
8. If 𝑨 = (−𝟒 𝟓 𝟐) , 𝒃 = (𝟎) ,
𝟔 −𝟗 𝟏 𝟔
solve the equation 𝑨𝒙 = 𝒃 by using the LU factorization.

2 −4 2 1 0 0 2 −4 2
(−4 5 2) = (−2 1 0) (0 𝑥 𝑦)
6 −9 1 3 𝑎 1 0 0 𝑧

2 −4 2 2 −4 2
(−4 5 2) = (−4 8+𝑥 −4 + 𝑦 )
6 −9 1 6 −12 + 𝑎𝑥 6 + 𝑎𝑦 + 𝑧

Equating components, 𝑥 = −3, 𝑦 = 6, 𝑎𝑥 = 3, 𝑧 = −5 − 𝑎𝑦

Therefore, 𝑥 = −3, 𝑦 = 6, 𝑎 = −1, 𝑧 = 1

1 0 0 2 −4 2
Therefore, 𝑳 = (−2 1 0) and 𝑈 = (0 −3 6)
3 −1 1 0 0 1

𝐴𝑥 = 𝑏
(𝐿𝑈)𝑥 = 𝑏
𝐿(𝑈𝑥) = 𝑏
𝐿𝑦 = 𝑏, where 𝑦 = 𝑈𝑥
1 0 0 𝑦1 6
(−2 1 0) (𝑦2 ) = (0)
3 −1 1 𝑦3 6
𝑦1 = 6
−2𝑦1 + 𝑦2 = 0
3𝑦1 − 𝑦2 + 𝑦3 = 6
By solving, 𝑦1 = 6, 𝑦2 = 12, 𝑦3 = 0
𝑈𝑥 = 𝑦
2 −4 2 𝑥1 6
(0 𝑥
−3 6) ( 2 ) = (12)
0 0 1 𝑥3 0
2𝑥1 − 4𝑥2 + 2𝑥3 = 6
−3𝑥2 + 6𝑥3 = 12
𝑥3 = 0
By solving, 𝑥1 = −5, 𝑥2 = −4, 𝑥3 = 0
−5
The solution is 𝑥 = (−4)
0

Dr. Narasimhan G, RNSIT 17

 𝟏 −𝟏 𝟐 𝟎
9. If 𝑨 = (𝟏 −𝟑 𝟏) , 𝒃 = (−𝟓),
𝟑 𝟕 𝟓 𝟕
solve the equation 𝑨𝒙 = 𝒃 by using the LU factorization.

1 −1 2 1 0 0 1 −1 2
(1 −3 1) = (1 1 0) (0 𝑥 𝑦)
3 7 5 3 𝑎 1 0 0 𝑧

1 −1 2 1 −1 2
(1 −3 1) = (1 −1 + 𝑥 2+𝑦 )
3 7 5 3 −3 + 𝑎𝑥 6 + 𝑎𝑦 + 𝑧

Equating components, 𝑥 = −2, 𝑦 = −1, 𝑎𝑥 = 10, 𝑧 = −1 − 𝑎𝑦

Therefore, 𝑥 = −2, 𝑦 = −1, 𝑎 = −5, 𝑧 = −6

1 0 0 1 −1 2
Therefore, 𝑳 = (1 1 0) and 𝑈 = (0 −2 −1)
3 −5 1 0 0 −6

𝐴𝑥 = 𝑏
(𝐿𝑈)𝑥 = 𝑏
𝐿(𝑈𝑥) = 𝑏
𝐿𝑦 = 𝑏, where 𝑦 = 𝑈𝑥
1 0 0 𝑦1 0
(1 1 0) (𝑦2 ) = (−5)
3 −5 1 𝑦3 7
𝑦1 = 0
𝑦1 + 𝑦2 = −5
3𝑦1 − 5𝑦2 + 𝑦3 = 7
By solving, 𝑦1 = 0, 𝑦2 = −5, 𝑦3 = −18
𝑈𝑥 = 𝑦
1 −1 2 𝑥1 0
(0 𝑥
−2 −1) ( 2 ) = ( −5 )
0 0 −6 𝑥3 −18
𝑥1 − 𝑥2 + 2𝑥3 = 0
−2𝑥2 − 𝑥3 = −5
−6𝑥3 = −18
By solving, 𝑥1 = −5, 𝑥2 = 1, 𝑥3 = 3
−5
The solution is 𝑥 = ( 1 )
3

Dr. Narasimhan G, RNSIT 18

 𝟏 𝟎
10. Find 𝑬𝟐 , 𝑬𝟖 and 𝑬−𝟏 for ( )
𝟔 𝟏

1 0 0 0
𝐸=( )+( )=𝐼+𝑁
0 1 6 0
𝐸 𝑘 = (𝐼 + 𝑁)𝐾 = 𝐼 + 𝑘𝑁, 𝑠𝑖𝑛𝑐𝑒 𝑁 2 = 0.
1 0 0 0 1 0
𝐸 2 = (𝐼 + 𝑁)2 = 𝐼 + 2𝑁 = ( ) + 2( )=( )
0 1 6 0 12 1
1 0 0 0 1 0
𝐸 8 = (𝐼 + 𝑁)8 = 𝐼 + 8𝑁 = ( ) + 8( )=( )
0 1 6 0 48 1
1 0 0 0 1 0
𝐸 −1 = (𝐼 + 𝑁)−1 = 𝐼 − 𝑁 = ( )−( )=( )
0 1 6 0 −6 1

Dr. Narasimhan G, RNSIT 19

 1.3 Vector spaces and subspaces

Definition: Vector spaces

A non-empty set of vectors V over the scalar field F under addition and scalar multiplication
is said to be a vector space if it satisfies the following properties:
If 𝑢, 𝑣, 𝑤 ∈ 𝑉 and 𝛼, 𝛽 ∈ 𝐹 then
Addition
(i) Closure: 𝑢 + 𝑣 ∈ 𝑉
(ii) Associative: (𝑢 + 𝑣) + 𝑤 = 𝑢 + (𝑣 + 𝑤)
(iii) Identity: There is 0 ∈ 𝑉 such that 0 + 𝑢 = 𝑢 = 𝑢 + 0
(iv) Inverse: There is −𝑢 ∈ 𝑉 such that 𝑢 + (−𝑢) = (−𝑢) + 𝑢 = 0
(v) Commutative: 𝑢 + 𝑣 = 𝑣 + 𝑢
Scalar multiplication
(vi) Closure: 𝛼𝑢 ∈ 𝑉
(vii) Distributive 1: 𝛼 (𝑢 + 𝑣) = 𝛼𝑢 + 𝛼𝑣
(viii) Distributive 2: (𝛼 + 𝛽) 𝑢 = 𝛼𝑢 + 𝛽𝑢
(ix) Associative: 𝛼(𝛽𝑢) = (𝛼𝛽)𝑢
(x) Identity: 1𝑢 = 𝑢
Example:
1. Let 𝑽 = {(𝒙, 𝒚) ∈ 𝑹𝟐 |𝒙𝟐 + 𝒚𝟐 ≤ 𝟏} be the unit disc in 𝑹𝟐 .
(1, 0) ∈ 𝑉 , but 2(1, 0) = (2, 0) ∉ 𝑉
The circle is not closed under scalar multiplication.
Therefore, the unit disc is not a vector space.
2. Let 𝑽 = {(𝒙, 𝒚) ∈ 𝑹𝟐 | 𝒚 = 𝒙𝟐 } be the graph of the quadratic function.
(1, 1), (2,4) ∈ 𝑉. But (1, 1) + (2,4) = (3, 5) ∉ 𝑉.
Therefore, V is not closed under addition.
Therefore, the graph of the quadratic function is not a vector space.
3. Let 𝑽 = {(𝒙, 𝒚) ∈ 𝑹𝟐 | 𝒚 = 𝟐𝒙} be the graph of the function 𝒇(𝒙) = 𝟐𝒙.
All the axioms of vector space are satisfied.
Therefore, the graph of the function f(x) = 2x is a vector space.
4. Let V be the set of all 𝟐 × 𝟐 matrices .
All the axioms of vector space are satisfied.
Therefore, V is a vector space.

Dr. Narasimhan G, RNSIT 20

 Definition: Subspaces

A subset W of a vector space 𝑉 is called a subspace of 𝑉 if it satisfies the properties

(i) The zero vector of 𝑉 is also in 𝑊.
(ii) If 𝑢, 𝑣 ∈ 𝑊 then 𝑢 + 𝑣 ∈ 𝑊 and 𝑘𝑢 ∈ 𝑊 for some scalar k.
Note:
❖ W is closed under addition and scalar multiplication.
❖ Suppose W is a subspace of V.
𝑢, 𝑣 ∈ 𝑊 ⟹ 𝛼𝑢 + 𝛽𝑣 ∈ 𝑊, for scalars 𝛼 and 𝛽.
❖ Suppose 𝑊1 , 𝑊2 be the subspaces of V.
Then 𝑊1 ∩ 𝑊2 is a subspace of 𝑉. But 𝑊1 ∪ 𝑊2 need not be the subspace of 𝑉.
Example:
1. If 𝑾 = {(𝒙, 𝒚) ∈ 𝑹𝟐 | 𝒚 = 𝟐𝒙} 𝒊𝒔 the graph of the function 𝒇(𝒙) = 𝟐𝒙 , then W
is a subspace of 𝑽 = 𝑹𝟐 .

(0, 2(0)) = (0, 0) ∈ 𝑊. Let (𝑎, 2𝑎), (𝑏, 2𝑏) ∈ 𝑊.

(𝑎, 2𝑎) + (𝑏, 2𝑏) = (𝑎 + 𝑏, 2𝑎 + 2𝑏) = (𝑎 + 𝑏, 2 (𝑎 + 𝑏)) ∈ 𝑊
𝑘(𝑎, 2𝑎) = (𝑘𝑎, 2(𝑘𝑎)) ∈ 𝑊 for some scalar 𝑘.
Therefore, W is closed under addition and scalar multiplication.
Therefore, W is a subspace of V.

2. Let 𝑾 = {(𝒙, 𝒚) ∈ 𝑹𝟐 | 𝒙 ≥ 𝟎, 𝒚 ≥ 𝟎} be the first quadrant in 𝑹𝟐 . W is not the

subspace of V.
Clearly (0, 0) ∈ 𝑊. Let (𝑥1 , 𝑦1 ), (𝑥2 , 𝑦2 ) ∈ 𝑊, where 𝑥𝑖 are non-negative components.
(𝑥1 + 𝑥2 , 𝑦1 + 𝑦2 ) ∈ 𝑊 , where all are non-negative components.
(1, 1) ∈ 𝑊 , but (−1)(1, 1) = (−1, −1) ∉ 𝑊
Therefore, 𝑊 is closed under addition, but not closed under scalar multiplication.
Therefore, W is not a subspace of V.

Dr. Narasimhan G, RNSIT 21

 3. Consider the subset W of 𝑽 = 𝑹𝟑 consisting of vectors of the form (𝒂, 𝒂, 𝒃),
where the first two components are the same. Then W is a subspace of V.

Clearly (0, 0, 0) ∈ 𝑊. Let (𝑎, 𝑎, 𝑏), (𝑐, 𝑐, 𝑑) ∈ 𝑊

(𝑎, 𝑎, 𝑏) + (𝑐, 𝑐, 𝑑) = (𝑎 + 𝑐, 𝑎 + 𝑐, 𝑏 + 𝑑) ∈ 𝑊
𝑘(𝑎, 𝑎, 𝑏) = (𝑘𝑎, 𝑘𝑎, 𝑘𝑏) ∈ 𝑊.
Thus, W is closed under addition and scalar multiplication.
Therefore, W is a subspace of V.

4. Consider the subset W of V consisting of vectors of the form (𝒂, 𝒂𝟐 , 𝒃), where the
second component is the square of the first. Then W is a not a subspace of V.

Clearly (0, 0, 0) ∈ 𝑊. Let (𝑎, 𝑎2 , 𝑏) , (𝑐, 𝑐 2 , 𝑑) ∈ 𝑊

(𝑎, 𝑎2 , 𝑏) + (𝑐, 𝑐 2 , 𝑑) = (𝑎 + 𝑐, 𝑎2 + 𝑐 2 , 𝑏 + 𝑑) ∉ 𝑊. [∵ 𝑎2 + 𝑐 2 ≠ (𝑎 + 𝑐)2 ]
Thus, W is not closed under addition.
Therefore, W is not a subspace of V.

5. Set W of 𝟐 × 𝟐 diagonal matrices is a subspace of the vector space 𝑴𝟐𝟐 .

0 0 𝑎 0 𝑐 0
Clearly ( ) ∈ 𝑊 . Let ( ), ( )∈𝑊
0 0 0 𝑏 0 𝑑
𝑎 0 𝑐 0 𝑎+𝑐 𝑜
( )+ ( )=( )∈𝑊
0 𝑏 0 𝑑 0 𝑏 + 𝑑
𝑎 0 𝑘𝑎 0
𝑘( )=( ) ∈ 𝑊.
0 𝑏 0 𝑘𝑏
Therefore, 𝑊 is closed under addition and scalar multiplication.
Therefore, W is a subspace of V.

Dr. Narasimhan G, RNSIT 22

 1.4 Linearly independent, Basis and dimension
Linear independence:

Let V be a vector over the field F. The vectors 𝑣1 , 𝑣2 , … , 𝑣𝑛 ∈ 𝑉 are said to be linearly
independent over F if there exist scalars 𝑎1 , 𝑎2 , … , 𝑎𝑛 ∈ 𝐹 such that
𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 = 0 ⇒ 𝑎1 = 0, 𝑎2 = 0, … , 𝑎𝑛 = 0. Otherwise, they are
linearly dependent.

Basis of a vector space:

The set of vectors {𝑣1 , 𝑣2 , … , 𝑣𝑛 } is called a basis of the vector space V, if
(i) 𝑣1 , 𝑣2 , … , 𝑣𝑛 are linearly independent (ii) 𝑣1 , 𝑣2 , … , 𝑣𝑛 span V. That is, each vector of V
can be uniquely expressed as linear combination of 𝑣1 , 𝑣2 , … , 𝑣𝑛 .

Dimension of a vector space V:

Number of elements in a basis of vector space V is called the dimension of V.

It is denoted by dim 𝑉. 𝐼𝑓 𝑉 contains a basis with n elements then the dim 𝑉 = 𝑛.

Note:

(i) If 𝑉 = {0} then dim 𝑉 = 0.

(ii) If dim 𝑉 is not finite then it is said to be of infinite dimension.

1. Check whether the vectors 𝒗𝟏 = (𝟏, 𝟐, 𝟑), 𝒗𝟐 = (𝟑, 𝟏, 𝟕) and 𝒗𝟑 = (𝟐, 𝟓, 𝟖) are
linearly dependent or not.
𝑥𝑣1 + 𝑦𝑣2 + 𝑧𝑣3 = 0
⟹ 𝑥(1, 2, 3) + 𝑦(3, 1, 7) + 𝑧(2, 5, 8) = (0, 0, 0)

⟹ 𝑥 + 3𝑦 + 2𝑧 = 0, 2𝑥 + 𝑦 + 5𝑧 = 0, 3𝑥 + 7𝑦 + 8𝑧 = 0

⟹ (𝑥, 𝑦, 𝑧) = (0, 0, 0)
Therefore, the given vectors are linearly independent.
2. Check whether the vectors 𝒗𝟏 = (𝟏, 𝟒, 𝟗), 𝒗𝟐 = (𝟑, 𝟏, 𝟒) and 𝒗𝟑 = (𝟗, 𝟑, 𝟏𝟐) are
linearly dependent or not.
𝑥𝑣1 + 𝑦𝑣2 + 𝑧𝑣3 = 0
⟹ 𝑥(1, 4, 9) + 𝑦(3, 1, 4) + 𝑧(9, 3, 12) = 0
⟹ 𝑥 + 3𝑦 + 9𝑧 = 0, 4𝑥 + 𝑦 + 3𝑧 = 0, 9𝑥 + 4𝑦 + 12𝑧 = 0
⟹ (𝑥, 𝑦, 𝑧) ≠ (0, 0, 0)
Therefore, the given vectors are linearly dependent.

Dr. Narasimhan G, RNSIT 23

 3. If 𝒖, 𝒗, 𝒘 are linearly independent vectors in V(F), where F is the field of Complex
numbers, then {𝒖 + 𝒗, 𝒗 + 𝒘, 𝒘 + 𝒖} is a linearly independent set of vectors.
Let 𝑎(𝑢 + 𝑣) + 𝑏(𝑣 + 𝑤) + 𝑐(𝑤 + 𝑢) = 0, 𝑤ℎ𝑒𝑟𝑒 𝑎, 𝑏, 𝑐 ∈ 𝐹
⟹ (𝑎 + 𝑐)𝑢 + (𝑎 + 𝑏)𝑣 + (𝑏 + 𝑐)𝑤 = 0
Since 𝑢, 𝑣, 𝑤 are linearly independent, 𝑎 + 𝑐 = 0, 𝑎 + 𝑏 = 0, 𝑏 + 𝑐 = 0.
⟹ 𝑎 = 0, 𝑏 = 0, 𝑐 = 0.
Therefore, {𝑢 + 𝑣, 𝑣 + 𝑤, 𝑤 + 𝑢} is a linearly independent set of vectors.

4. Let V be a vector space of all 𝟐 × 𝟑 matrices over R. Show that the matrix
𝟐 𝟏 −𝟏 𝟏 𝟏 −𝟑 𝟒 −𝟏 𝟐
𝑨=[ ], 𝑩 = [ ], 𝑪 = [ ] form a linearly
𝟑 −𝟐 𝟒 𝟐 𝟎 𝟓 𝟏 −𝟐 𝟑
independent set.
2 1 −1 1 1 −3 4 −1 2 0 0 0
Let 𝑎 [ ]+𝑏[ ]+𝑐[ ]=[ ]
3 −2 4 2 0 5 1 −2 3 0 0 0
2𝑎 + 𝑏 + 4𝑐 𝑎 + 𝑏 − 𝑐 −𝑎 − 3𝑏 + 2𝑐 0 0 0
[ ]=[ ]
3𝑎 − 2𝑏 + 𝑐 −2𝑎 − 2𝑐 4𝑎 + 5𝑏 + 3𝑐 0 0 0
By equating the corresponding elements,
2𝑎 + 𝑏 + 4𝑐 = 0, 𝑎 + 𝑏 − 𝑐 = 0, −𝑎 − 3𝑏 + 2𝑐 = 0. ---- (1)
3𝑎 − 2𝑏 + 𝑐 = 0, −2𝑎 − 2𝑐 = 0, 4𝑎 + 5𝑏 + 3𝑐 = 0. ----- (2)
On solving the system (1) , 𝑎 = 0, 𝑏 = 0, 𝑐 = 0.
This solution also satisfies the system (2).
Therefore, the given set of matrices is linearly independent.

5. Determine whether 𝒙𝟏 = (𝟐, 𝟐, 𝟏), 𝒙𝟐 = (𝟏, 𝟑, 𝟕), 𝒙𝟑 = (𝟏, 𝟐, 𝟐) in 𝑹𝟑 form a basis.

Three vectors in 𝑅 3 form a basis if and only if they are linearly independent.
𝑥(2, 2, 1) + 𝑦(1, 3, 7) + 𝑧(1, 2, 2) = (0, 0, 0)
⟹ 2𝑥 + 𝑦 + 𝑧 = 0, 2𝑥 + 3𝑦 + 2𝑧 = 0, 𝑥 + 7𝑦 + 2𝑧 = 0
⟹ (𝑥, 𝑦, 𝑧) = (0, 0, 0)
Therefore, the vectors 𝑥1 , 𝑥2 , 𝑥3 are linearly independent and hence form a basis.

Dr. Narasimhan G, RNSIT 24

 6. Let W be the subspace of 𝑹𝟓 spanned by 𝒙𝟏 = (𝟏, 𝟐, −𝟏, 𝟑, 𝟒),
𝒙𝟐 = (𝟐, 𝟒, −𝟐, 𝟔, 𝟖), 𝒙𝟑 = (𝟏, 𝟑, 𝟐, 𝟐, 𝟔), 𝒙𝟒 = (𝟏, 𝟒, 𝟓, 𝟏, 𝟖), 𝒙𝟓 = (𝟐, 𝟕, 𝟑, 𝟑, 𝟗).
Find a subset of vectors which forms a basis of W.
1 2 −1 3 4
2 4 −2 6 8
𝐴= 1 3 2 2 6
1 4 5 1 8
[2 7 3 3 9]
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 𝑅1 , 𝑅4 → 𝑅4 − 𝑅1 , 𝑅5 → 𝑅5 − 𝑅1
1 2 −1 3 4
0 0 0 0 0
∼ 0 1 3 −1 2
0 2 6 −2 4
[0 3 5 −3 1]
𝑅2 ↔ 𝑅5
1 2 −1 3 4
0 3 5 −3 1
∼ 0 1 3 −1 2
0 2 6 −2 4
[0 0 0 0 0]
𝑅2 ↔ 𝑅3
1 2 −1 3 4
0 1 3 −1 2
∼ 0 3 5 −3 1
0 2 6 −2 4
[0 0 0 0 0]
𝑅3 → 𝑅3 − 3𝑅2 , 𝑅4 → 𝑅4 − 2𝑅2
1 2 −1 3 4
0 1 3 −1 2
∼ 0 0 −4 0 −5
0 0 0 0 0
[0 0 0 0 0]
The number of non-zero rows is 3.
Therefore, dim 𝑊 = 3 and {𝑥1 , 𝑥2 , 𝑥3 } forms a basis in W.

Dr. Narasimhan G, RNSIT 25

 7. V is a vector space of polynomials over R. Find a basis and dimension of the
subspace W of V, spanned by the polynomials, 𝒙𝟏 = 𝒕𝟑 − 𝟐𝒕𝟐 + 𝟒𝒕 + 𝟏,
𝒙𝟐 = 𝟐𝒕𝟑 − 𝟑𝒕𝟐 + 𝟗𝒕 − 𝟏, 𝒙𝟑 = 𝒕𝟑 + 𝟔𝒕 − 𝟓, 𝒙𝟒 = 𝟐𝒕𝟑 − 𝟓𝒕𝟐 + 𝟕𝒕 + 𝟓.

1 −2 4 1
2 −3 9 −1
𝐴=[ ]
1 0 6 −5
2 −5 7 5
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 𝑅1 , 𝑅4 → 𝑅4 − 2𝑅1
1 −2 4 1
0 1 1 −3
∼[ ]
0 2 2 −6
1 −1 −1 3
𝑅3 → 𝑅3 − 2𝑅2 , 𝑅4 → 𝑅4 + 𝑅2
1 −2 4 1
0 1 1 −3
∼[ ]
0 0 0 0
0 0 0 0
Non-zero rows of Echelon matrix form a basis.
Number of non-zero rows is 2.
Therefore, dim 𝑊 = 2 and {𝑥1 , 𝑥2 } forms a basis of W.

Dr. Narasimhan G, RNSIT 26

 1.5 Row space and column space
Introduction:
❖ The first non-zero number in each row of the echelon form of the given matrix is called
pivot number and its corresponding column is called pivot column.
❖ The column space of A is the span of the pivot columns of A .
The basis of the column space is the set of all pivot columns of A.
❖ The Row space of 𝐴 is the span of the non-zero rows in the Echelon form of A.
The basis of Row space of 𝐴 is the set of all non-zero rows in the Echelon form of A.
❖ Null space of A = {𝑥|𝐴𝑥 = 0}

Definition:
❖ The column space of A is the set of all possible linear combinations of its column
vectors. It is denoted by 𝐶(𝐴). Its dimension of 𝐶(𝐴) is 𝑟. This is also called as the
range.
❖ The row space of 𝐴 is the set of all possible linear combinations of its row vectors.
It is the column space of 𝐴𝑇 . It is denoted by 𝐶(𝐴𝑇 ). It is spanned by the rows of A.
Its dimension is r.

𝟏 𝟐
1. What is the column space of ( ) ? Also find the basis and the dimension of the
𝟑 𝟔
column space of the given matrix.
1 2
Let 𝐴 = ( )
3 6
𝑅2 → 𝑅2 − 3𝑅1
The row reduced Echelon form of A is
1 2
𝑈=( ).
0 0
First column is the pivot column.
1
Basis of the column space = {( )}
3
Dimension of the column space = 1.
1
The column space of A = 𝑠𝑝𝑎𝑛 {( )}
3

Dr. Narasimhan G, RNSIT 27

 𝟏 𝟐
2. What is the row space of ( ) ? Also find the basis and the dimension of the row
𝟑 𝟔
space of the given matrix.

1 2
Let 𝐴 = ( )
3 6

𝑅2 → 𝑅2 − 3𝑅1

The row reduced Echelon form of A is

1 2
𝑈=( ).
0 0

Basis of the row space = {(1, 2)}

Dimension of the row space = 1
The row space = 𝑠𝑝𝑎𝑛 {(1, 2)}

𝟏 𝟏 −𝟑 𝟐
3. Find the row space, column space of ( )
𝟑 𝟒 −𝟏𝟏 𝟕

1 1 −3 2
𝐴=( )
3 4 −11 7
𝑅2 → 𝑅2 − 3𝑅2
1 1 −3 2
𝑈=( )
0 1 −2 1
This is the row reduced echelon form of A.
The two rows are the non-zero rows.
Basis of row space = {(1, 1, −3, 2), (0, 1, 1, −2, 1)}
Dimension of the row space = 2
Row space = 𝑠𝑝𝑎𝑛 {(1, 1, −3, 2), (0, 1, 1, −2, 1)}
The first column is the pivot column.
1
Basis of column space = {( )}
3
Dimension of the column space is 1.
1
Column space = 𝑠𝑝𝑎𝑛 {( )}
3

Dr. Narasimhan G, RNSIT 28

 𝟏 𝟐 𝟑 𝟓
4. Find the row space, column space of 𝑨 = (𝟐 𝟒 𝟖 𝟏𝟐)
𝟑 𝟔 𝟕 𝟏𝟑
1 2 3 5
𝑨 = (2 4 8 12)
3 6 7 13
𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 3𝑅1
1 2 3 5
~ (0 0 2 2)
0 0 −2 −2
𝑅3 → 𝑅3 + 𝑅2
𝟏 2 3 5
𝑈 = (0 0 𝟐 2)
0 0 0 0
This is the row reduced echelon form of A.
The first two rows are the non-zero rows.
Basis of row space = {(1, 2, 3, 5), (0, 0, 2, 2)}
Dimension of the row space = 2
Row space = 𝑠𝑝𝑎𝑛 {(1, 2, 3, 5), (0, 0, 2, 2)}
1 3
Basis of column space = {(2) , (8)}
3 7
Dimension of the column space is 2.
1 3
Column space = 𝑠𝑝𝑎𝑛 {(2) , (8)}
3 7

Dr. Narasimhan G, RNSIT 29

 −𝟐 −𝟓 𝟖 𝟎 −𝟏𝟕
𝟏 𝟑 −𝟓 𝟏 𝟓
5. Find the row space and column space of the matrix ( )
𝟑 𝟏𝟏 −𝟏𝟗 𝟕 𝟏
𝟏 𝟕 −𝟏𝟑 𝟓 −𝟑
−2 −5 8 0 −17
1 3 −5 1 5
𝐴=( )
3 11 −19 7 1
1 7 −13 5 −3
𝑅1 ↔ 𝑅2
1 3 −5 1 5
−2 −5 8 0 −17
~( )
3 11 −19 7 1
1 7 −13 5 −3
𝑅2 → 𝑅2 + 2𝑅1 , 𝑅3 → 𝑅3 − 3𝑅1 , 𝑅4 → 𝑅4 − 𝑅1
1 3 −5 1 5
0 1 −2 2 −7
~( )
0 −22 −4 4 −14
0 4 −8 4 −8
𝑅3 → 𝑅3 + 22𝑅2 , 𝑅4 → 𝑅4 − 4𝑅1
1 3 −5 1 5
0 1 −2 2 −7
~( )
0 0 −48 48 −168
0 0 0 −4 20
𝑅3 → 𝑅3 − 24𝑅1
1 3 −5 1 5
0 1 −2 2 −7
~( )
0 0 0 0 0
0 0 0 −4 20
𝑅3 ↔ 𝑅4
1 3 −5 1 5
0 1 −2 2 −7
~( )
0 0 0 −4 20
0 0 0 0 0
𝑅3
𝑅3 →
−4

1 3 −5 1 5
0 1 −2 2 −7
𝑈=( )
0 0 0 1 −5
0 0 0 0 0

This is the echelon form of A.

The non-zero rows are the first three rows.
Basis of row space = {(1, 3, −5, 1, 5), (0, 1, −2, 2, −7), (0, 0, 0, 1, −5)}

Dr. Narasimhan G, RNSIT 30

 Dimension of the row space is 3.
Row space = 𝑠𝑝𝑎𝑛 {(1, 3, −5, 1, 5), (0, 1, −2, 2, −7), (0, 0, 0, 1, −5)}
First, second and fourth columns are pivot columns.
−2 −5 0
1 3 1
Basis of column space = {( ) , ( ) , ( )}
3 11 7
1 7 5
Dimension of the column space is 3.

−2 −5 0
1 3 1
Column space = 𝑠𝑝𝑎𝑛 {( ) , ( ) , ( )}
3 11 7
1 7 5

Dr. Narasimhan G, RNSIT 31

 1.6 The four fundamental subspaces

Subspaces Notation Dimension

Column space 𝐶(𝐴) 𝑟

Null space 𝑁(𝐴) 𝑛−𝑟

Row space 𝐶(𝐴𝑇 ) 𝑟

Left null space 𝑁(𝐴𝑇 ) 𝑚−𝑟

1. The column space of A is the set of all possible linear combinations of its column
vectors. It is denoted by 𝐶(𝐴). Its dimension of 𝐶(𝐴) is 𝑟. This is also called as the
range.

2. The row space of 𝐴 is the set of all possible linear combinations of its row vectors.
It is the column space of 𝐴𝑇 . It is denoted by 𝐶(𝐴𝑇 ). It is spanned by the rows of A.
Its dimension is r.

3. The null space of 𝐴 is denoted by 𝑁(𝐴). Its dimension is 𝑛 − 𝑟. 𝑁(𝐴) = {𝑋|𝐴𝑋 = 0}.
This is also known as kernel of A. Its dimension is nullity.

4. The left null space of A is the null space of 𝐴𝑇 . It contains all vectors 𝑦 such that
𝐴𝑇 𝑦 = 0. It is denoted by 𝑁(𝐴𝑇 ). Its dimension is 𝑚 − 𝑟. 𝑁(𝐴𝑇 ) = {𝑌|𝐴𝑇 𝑌 = 0}

Dr. Narasimhan G, RNSIT 32

 𝟏 𝟐
1. Find the null space and left null space of 𝑨 = ( ).
𝟑 𝟔

To find: Null space To find: Left null space

1 2 1 3
𝐴=( ) 𝐴𝑇 = ( )
3 6 2 6
RREF of A RREF of A
1 3
𝑈=(
1 2
) 𝑈𝑇 = ( )
0 0 0 0
𝐴𝑋 = 0 𝐴𝑇 𝑌 = 0

𝑈𝑋 = 0 𝑈𝑇 𝑌 = 0

1 2 𝑥1 0 1 3 𝑦1 0
( )( ) = ( ) ( )( ) = ( )
0 0 𝑥2 0 0 0 𝑦2 0
𝑥2 is the free variable say 𝑘. 𝑦2 is the free variable say 𝑘.

𝑥1 + 2𝑥2 = 0 𝑦1 + 3𝑦2 = 0

𝑥1 = −2𝑘 𝑦1 = −3𝑘
𝑥1 𝑥2 𝑦1 −3𝑘 −3
= (𝑦 ) = ( ) = 𝑘( )
−2 1 2 𝑘 1
𝑥1 −3
(𝑥 ) = (
−2𝑘 −2
) = 𝑘( ) 𝑁(𝐴𝑇 ) = span {( )}
2 𝑘 1 1
−2
𝑁(𝐴) =span {( )}
1

Dr. Narasimhan G, RNSIT 33

 𝟏 𝟐 𝟑 𝟓
2. Find the null space and left null space of 𝑨 = (𝟐 𝟒 𝟖 𝟏𝟐)
𝟑 𝟔 𝟕 𝟏𝟑

To find: Null space To find: Left null space

1 2 3 5 1 2 3
𝐴 = (2 4 8 12) 2 4 6
𝐴𝑇 = ( )
3 6 7 13 3 8 7
5 12 13
RREF of A RREF of 𝐴𝑇
1 2 3 5 1 2 3
𝑈 = (0 0 2 2 ) 0 2 −2
𝑈𝑇 = ( )
0 0 0 0 0 0 0
𝐴𝑋 = 0 0 0 0
𝑈𝑋 = 0 𝐴𝑇 𝑌 = 0
𝑥 0
1 2 3 5 𝑥12 0 𝑈𝑇 𝑌 = 0
(0 0 2 2 ) ( 𝑥 ) = ( )
3 0
0 0 0 0 𝑥4 1 2 3
0 𝑦1 0
𝑥1 , 𝑥3 are basic variables 0 2 −2
( ) (𝑦2 ) = (0)
0 0 0 𝑦3
𝑥2 , 𝑥4 are free variables say 𝑝, 𝑞 0 0 0 0
𝑥1 + 2𝑥2 + 3𝑥3 + 5𝑥4 = 0
𝑦3 is the free variable. Let 𝑦3 = 𝑘
and 2𝑥3 + 2𝑥4 = 0
⟹ 𝑥3 = −𝑞, 𝑥1 = −2𝑝 − 2𝑞 𝑦1 + 2𝑦2 + 3𝑦3 = 0, 2𝑦2 − 2𝑦3 = 0
Therefore,
𝑥1 −2𝑝 − 2𝑞 ⟹ 𝑦2 = 𝑘, 𝑦1 = −5𝑘
𝑥2 𝑝
(𝑥 ) = ( −𝑞 ) 𝑦1 −5𝑘 −5
3
𝑦
( 2) = ( 𝑘 ) = 𝑘 ( 1 )
𝑥4 𝑞
𝑦3 𝑘 1
−2 −2 −5
1 0
= 𝑝( )+𝑞( ) 𝑁(𝐴𝑇 ) = 𝑆𝑝𝑎𝑛 {( 1 )}
0 −1
1
0 1
−2 −2
1 0
𝑁(𝐴) = 𝑆𝑝𝑎𝑛 {( ) , ( )}
0 −1
0 1

Note:

Matrix, The null space 𝑁(𝐴) and row space 𝐶(𝐴𝑇 ) are subspaces of 𝑅 𝑛 .

The left null space 𝑁(𝐴𝑇 ) and column space 𝐶(𝐴) are subspaces of 𝑅 𝑚 .

Dr. Narasimhan G, RNSIT 34

 3. Find the dimension and construct a basis for the four sub spaces associated with
𝟎 𝟏 𝟒 𝟎
the matrix ( ).
𝟎 𝟐 𝟖 𝟎

0 1 4 0
Let 𝐴 = ( )
0 2 8 0
𝑅2 → 𝑅2 − 2𝑅1
Echelon form of A is
0 1 4 0
𝑈=( )
0 0 0 0
1
Basis for the column space is {( )} and the dimension is 1.
2
Basis for the row space is {(0, 1, 4, 0)} and the dimension is 1.
Null space: Left null space:
𝐴𝑋 = 0 0 0
𝑈𝑋 = 0 1 2
𝑥1 𝐴𝑇 = ( )
0 4 8
0 1 4 0 𝑥2 0 0 0
( ) (𝑥 ) = ( )
0 0 0 0 3 0 0 0
𝑥4 0 1 2
𝑥1 , 𝑥3 , 𝑥4 are free variables say 𝑝, 𝑞, 𝑟 𝑈𝑇 = ( )
0 0
𝑥2 + 4𝑥3 = 0 ⟹ 𝑥2 = −4𝑞 0 0
𝑥1 𝑝 𝑇
𝑥2 𝐴 𝑌=0
−4𝑞
(𝑥 ) = ( )
3 𝑞 𝑈𝑇 𝑌 = 0
𝑥4 𝑟
0 0
1 0 0 1 2 𝑦1 0
0 −4 0 ( ) (𝑦 ) = ( )
= 𝑝( ) +𝑞( )+𝑟( ) 0 0 2 0
0 1 0 0 0
0 0 1
Basis for the null space 𝑦2 is the free variable say 𝑘.
1 0 0 𝑦1 + 2𝑦2 = 0
0 −4 0
= {( ) , ( ) , ( ) } 𝑦1 = −2𝑘
0 1 0
0 0 1 𝑦1 −2𝑘 −2
𝑦 = (𝑦 ) = ( ) = 𝑘( )
Dimension = 3 . 2 𝑘 1
Basis for the left null space
−2
= {( )}
1
Dimension = 1.

Dr. Narasimhan G, RNSIT 35

 4. Find the dimension and construct a basis for the four sub spaces associated with
1 2 0 1
the matrix (0 1 1 0)
1 2 0 1
1 2 0 1
Let 𝐴 = (0 1 1 0), 𝑅3 → 𝑅3 − 𝑅1
1 2 0 1
1 2 0 1
Echelon form of A is 𝑈 = (0 1 1 0)
0 0 0 0
1 2
Basis for the column space is {(0) , (1)} and the dimension is 2.
1 2

Basis for the row space is {(1, 2, 0, 1), (0, 1, 1, 0)} and the dimension is 2.

Null space: Left null space:

1 2 0 1 1 0 0
𝐴 = (0 1 1 0) 2 1 0
1 2 0 1 𝐴𝑇 = ( )
0 1 0
1 2 0 1 1 0 0
𝑈 = (0 1 1 0)
1 0 0
0 0 0 0
0 1 0
𝐴𝑋 = 0 𝑈𝑇 = ( )
0 0 0
𝑈𝑋 = 0 0 0 0
𝑥1 0 𝐴𝑇 𝑌 = 0
1 2 0 1 𝑥2 0 𝑈𝑇 𝑌 = 0
(0 1 1 0) (𝑥 ) = ( )
3 0
0 0 0 0 𝑥4 1 0 0 𝑦1
0 0
0 1 0
𝑥3 , 𝑥4 are free variables say 𝑝, 𝑞 ( ) (𝑦2 ) = (0)
0 0 0 𝑦3 0
𝑥1 + 2𝑥2 + 𝑥4 = 0, 𝑥2 + 2𝑥3 = 0 0 0 0

𝑥2 = −2𝑝, 𝑥1 = 4𝑝 − 𝑞 𝑦3 is the free variable say 𝑘.

𝑥1 4𝑝 − 𝑞 4 −1 𝑦1 = 0, 𝑦2 = 0
𝑥2 −2𝑝 −2 0 𝑦1 0 0
(𝑥 ) = ( ) = 𝑝( )+𝑞( ) 𝑦
3 𝑝 1 0 𝑦 = ( 2 ) = (0 ) = 𝑘 (0 )
𝑥4 𝑞 0 1 𝑦3 𝑘 𝑘
Basis for the null space Basis for the left null space
4 −1 0
−2 0 = {(0)}
= {( ) , ( ) } 1
1 0
0 1 Dimension = 1.
Dimension = 2 .

Dr. Narasimhan G, RNSIT 36

 5. Find the dimension and construct a basis for the four sub spaces associated with
0 1 0
the matrix (0 0 1)
0 0 0
0 1 0
Let 𝐴 = 𝑈 = (0 0 1)
0 0 0
1 0
Basis for the column space is {(0) , (1)} and the dimension is 2.
0 0

Basis for the row space is {(0, 1, 0), (0, 0, 1} and the dimension is 2.

Null space: Left null space:

0 1 0 0 0 0
𝐴 = 𝑈 = (0 0 1) 𝐴𝑇 = 𝑈 𝑇 = (1 0 0)
0 0 0 0 1 0
𝑈𝑋 = 0 𝑈𝑇 𝑌 = 0
0 1 0 𝑥1 0 0 0 0 𝑦1 0
(0 0 1) (𝑥2 ) = (0) (1 0 0) (𝑦2 ) = (0)
0 0 0 𝑥3 0 0 1 0 𝑦3 0
𝑥1 is the free variable say 𝑘. 𝑦3 is the free variable say 𝑘.
𝑥2 = 0, 𝑥3 = 0 𝑦1 = 0, 𝑦2 = 0
𝑥1 𝑘 1 𝑦1
𝑥 0 0
( 2 ) = (0) = 𝑘 (0)
𝑥3 𝑦 = (𝑦2 ) = (0) = 𝑘 (0)
0 0
𝑦3 𝑘 1
1
Basis for the null space = {(0) } 0
Basis for the left null space = {(0)}
0
1
Dimension = 1 .
Dimension = 1.

Dr. Narasimhan G, RNSIT 37

 6. Without multiplying matrices, find bases for the row and column spaces of A:
1 2
3 0 3
𝐴 = (4 5) ( )
1 1 2
2 7

1 2
3 0 3
𝐴 = (4 5) ( ) = 𝑃𝑄
1 1 2
2 7
1 2
Basis for the column space of A=Linearly independent columns of P= {(4) , (5)}
2 7

Basis for the row space of A = Linearly independent rows of Q = {(3, 0, 3), (1, 1, 2)}

Dr. Narasimhan G, RNSIT 38

 7. Without computing 𝑨 , find bases for the four fundamental subspaces :
𝟏 𝟎 𝟎 𝟏 𝟐 𝟑 𝟒
𝑨 = (𝟔 𝟏 𝟎 ) (𝟎 𝟏 𝟐 𝟑 )
𝟗 𝟖 𝟏 𝟎 𝟎 𝟏 𝟐
How do you know from these shapes that A is not invertible.
(i) Column space of A:
Since all the three columns of P are linearly independent,
Basis for the column space of A=Basis for the column space for P
1 0 0
= {(6) , (1) , (0) }
9 8 1
(ii) Row space of A:
Since all the three rows of Q are non-zero,
Basis for the row space of A = Basis for the row space for Q
= {(1, 2, 3, 4), (0, 1, 2, 3), (0, 0, 1, 2)}
(iii) Null space of A:
𝑥1 0
1 2 3 4 𝑥2 0
(0 1 2 3 ) ( 𝑥 ) = ( )
3 0
0 0 1 2 𝑥4 0
𝑥4 is the free variable, say s.

𝑥1 + 2𝑥2 + 3𝑥3 + 4𝑥4 = 0, 𝑥2 + 2𝑥3 + 3𝑥4 = 0, 𝑥3 + 2𝑥4 = 0

𝑥3 = −2𝑠, 𝑥2 = 𝑠, 𝑥1 = 0
𝑥1 0 0
𝑥2 𝑠 1
(𝑥 ) = ( ) = 𝑠( )
3 −2𝑠 −2
𝑥4 𝑠 1

0
1
Basis for the null space of A= {( )}
−2
1

(iv) Left null space of A:

1 0 0 𝑦1 0
(6 1 0) (𝑦2 ) = (0)
9 8 1 𝑦3 0

𝑦1 = 0, 6𝑦1 + 𝑦2 = 0, 9𝑦1 + 8𝑦2 + 𝑦3 = 0

𝑦1 0
(𝑦2 ) = (0)
𝑦3 0
0
Basis for the left null space of A= {(0)}
0

Dr. Narasimhan G, RNSIT 39

 8. Without elimination, find the dimension and bases for the four subspaces for 𝑨 =
𝟎 𝟑 𝟑 𝟑
(𝟎 𝟎 𝟎 𝟎).
𝟎 𝟏 𝟎 𝟏

(i) Column space of A:

Since the second and third columns of P are linearly independent,
3 3
Basis for the column space of A= {(0) , (0) }
1 0
(ii) Row space of A:
Since the first and third rows are non-zero,
Basis for the row space of A= {(0, 3, 3, 3), (0, 1, 0, 1)}
(iii) Null space of A:
𝑥1 0
0 3 3 3 𝑥2 0
(0 0 0 0 ) ( 𝑥 ) = ( )
3 0
0 1 0 1 𝑥4 0
𝑥1 , 𝑥4 is the free variables, say s and t.

3𝑥2 + 3𝑥3 + 3𝑥4 = 0, 𝑥2 + 𝑥4 = 0

𝑥1 = 𝑠, 𝑥2 = −𝑡, 𝑥3 = 0, 𝑥4 = 𝑡
𝑥1 𝑠 1 0 1 0
𝑥2 −𝑡 0 −1 0 −1
(𝑥 ) = ( ) = 𝑠 ( ) + 𝑡 ( ). Basis for the null space of A= {( ) , ( )}
3 0 0 0 0 0
𝑥4 𝑡 0 1 0 1

(iv) Left null space of A:

0 0 0 0 0 0
3 0 1 3 0 1
( )~ ( )
3 0 0 0 0 −1
3 0 1 0 0 0

0 0 0 𝑦1 0
3 0 1
( ) (𝑦2 ) = (0)
0 0 −1 𝑦3 0
0 0 0

𝑦2 is the free variable, say s.

3𝑦1 + 𝑦3 = 0, −𝑦3 = 0
𝑦3 = 0, 𝑦1 = 0
𝑦1 0 0 0
(𝑦2 ) = ( 𝑠 ) = 𝑠 (1). Basis for the left null space of A= {(1)}
𝑦3 0 0 0

Dr. Narasimhan G, RNSIT 40

 9. Without elimination, find the dimension and bases for the four subspaces for 𝑨 =
𝟏 𝟏
(𝟒 𝟒 ) .
𝟓 𝟓

(i) Column space of A:

Since both columns are linearly dependent,
1
Basis for the column space of A= {(4) }
5

(ii) Row space of A:

Since only the first row is linearly independent, ,
Basis for the row space of A= {(1, 1)}
(iii) Null space of A:
1 1 𝑥
1 0
(4 4) (𝑥 ) = ( )
2 0
5 5
𝑥2 is the free variable say s.
𝑥1 + 𝑥2 = 0, 𝑥1 = −𝑥2 = −𝑠
𝑥1 −𝑠 −1
(𝑥 ) = ( ) = 𝑠 ( )
2 𝑠 1
−1
Basis for the null space of A= {( )}
1

(iv) Left null space of A:

𝑦1 0
1 4 5 𝑦
( ) ( 2 ) = (0 )
1 4 5 𝑦
3 0
𝑦2 and 𝑦3 are free variables, say s, t.
𝑦1 + 4𝑦2 + 5𝑦3 = 0
𝑦1 = −4𝑠 − 5𝑡
𝑦1 −4𝑠 − 5𝑡 −4 −5
𝑦
( 2) = ( 𝑠 ) = 𝑠( 1 )+𝑡( 0 )
𝑦3 𝑡 0 1
−4 −5
Basis for the left null space of A= {( 1 ) , ( 0 )}
0 1

Dr. Narasimhan G, RNSIT 41

 𝟏 𝟎 𝟎 𝟎 𝟏 𝟐 𝟑 𝟒
10. Find a basis for each of the four subspaces of 𝑨 = (𝟏 𝟏 𝟎) (𝟎 𝟎 𝟎 𝟏 𝟐)
𝟎 𝟏 𝟏 𝟎 𝟎 𝟎 𝟎 𝟎
(i) Column space of A:
Since all the three columns of P are linearly independent,
Basis for the column space of A=Basis for the column space for P
1 0 0
= {(1 , 1 , 0) }
) ( ) (
0 1 1
(ii) Row space of A:
Since the first two rows of Q are non-zero,
Basis for the row space of A = Basis for the row space for Q
= {(0, 1, 2, 3, 4), (0, 0, 0, 1, 2)}
(iii) Null space of A:
𝑥1 0
0 1 2 3 4 𝑥2 0
(0 0 0 1 2) 𝑥3 = 0
0 0 0 0 0 𝑥4 0
(𝑥5 ) (0)
𝑥1 , 𝑥3 , 𝑥5 are the free variable, say 𝑝, 𝑞, 𝑟.

𝑥2 + 2𝑥3 + 3𝑥4 + 4𝑥5 = 0, 𝑥4 + 2𝑥5 = 0

𝑥2 = −2𝑞 + 2𝑟, 𝑥4 = −2𝑟

𝑥1 𝑝 1 0 0
𝑥2 −2𝑞 + 2𝑟 0 −2 2
𝑥3 = 𝑞 =𝑝 0 +𝑞 1 +𝑟 0
𝑥4 −2𝑟 0 0 −2
𝑥
( 5) ( 𝑟 ) (0) (0) (1)
1 0 0
0 −2 2
Basis for the null space of A= 0 , 1 , 0
0 0 −2
{(0) ( 0 ) ( 1 ) }

(iv) Left null space of A:

1 1 0 𝑦1 0
(0 1 1) (𝑦2 ) = (0).
0 0 1 𝑦3 0

𝑦1 + 𝑦2 = 0, 𝑦2 + 𝑦3 = 0, 𝑦3 = 0

0
Therefore, Basis for the left null space of A = {(0)}
0

Dr. Narasimhan G, RNSIT 42

 Note: 1
If 𝐴 = 𝑃𝑄, then
(i) Column space of A⊆ Column space of P
(ii) Row space of A ⊆ Row space of Q
(iii) Null space of Q ⊆ Null space of A
(iv) Left null space of P ⊆ Left null space of A
Note: 2
If 𝐴 = 𝐿𝑈, then
(i) Column space of A= Column space of P
(ii) Row space of A = Row space of Q
(iii) Null space of 𝐴 = Null space of Q
(iv) Left null space of A = Left null space of P

Dr. Narasimhan G, RNSIT 43