LINEAR ALGEBRA

BSCS (SEMESTER Ⅱ)
Lecture# 05
April 27, 2021
Section – E2
Amna Tahir
GCU, Lahore
Outline Special Forms of a Square Matrix

REF and RREF

Rank of a Matrix

2
 Special Forms of Matrices
Triangular Matrices:
▪ A square matrix is called upper triangular if all the entries below the main
diagonal are zero. It is typically denoted with U.

▪ A square matrix is called lower triangular if all the entries above the main
diagonal are zero. It is denoted with L.

▪ A matrix is triangular if it is either lower triangular or upper triangular.

▪ A matrix that is both upper and lower triangular is called a diagonal matrix.
𝟑 −8 3 𝟐 0 0 𝟐 0 0 2 0
0 𝟕 −3 0 𝟖 0 −1 𝟏𝟎 0 3 0
0 0 𝟏 0 0 𝟏 0 −2 𝟏 1 5
Upper Triangular Diagonal Lower Triangular Not Triangular

3
 Special Forms of Matrices
Echelon Form:
An 𝑚 × 𝑛 matrix 𝐴 is said to be in echelon form (row echelon form or REF) if it
has the following properties:
• All zero rows, if any, are at the bottom of the matrix.
• The first nonzero entry in each nonzero row (also called the leading entry) is to
the right of the first nonzero entry of each preceding row.
• There must be zeros below the leading entry.

Example: The following matrices are in echelon form;

𝟐 3 2
𝟑 −8 3 𝟏 −8 3 0 𝟏 3 0 0 4
0 0 𝟏
0 𝟕 −3 , 0 𝟏 − 26Τ9 , 0 0 0 , 0 0 0 𝟏 0 −3
0 0 𝟏 0 0 0 0 0 0 0 𝟏 2
0 0 0
4
 Special Forms of Matrices
Reduced Echelon Form:
An 𝑚 × 𝑛 matrix 𝐴 is said to be in reduced echelon form (reduced row echelon
form or RREF) if it has the following properties:
• 𝐴 is in echelon form.
• Each leading entry is equal to 1.
• Each leading entry is the only nonzero entry in its column (i.e. all entries above
and below the leading entry are zero).

Example: The following matrices are in reduced echelon form;

𝟏 0 3 𝟏 0 2 𝟏 2 0 7 0
0 𝟏 −3 , 0 𝟏 −1 , 0 0 𝟏 2 0
0 0 0 0 0 0 0 0 0 0 𝟏

5
Leading Entries Position (in REF & RREF)

6
 Some Examples (REF & RREF)
1 2 3 5 1 2 3 5
▪ 0 0 1 2 , 0 0 1 2 These matrices are not in echelon forms.
0 0 0 0 0 0 1 0
0 0 0 4 0 0 0 4

1 2 3 5
0 2 5 1
▪ 0 0 1 2 ,
0 0 9 −1 These matrices are in echelon form only.
0 0 0 4
0 0 0 0
0 0 0 0

1 2 0 5 1 3 0 0
▪ 0 0 1 2 , 0 0 1 0 These matrices are in both echelon & reduced
0 0 0 0 0 0 0 1
0 0 0 0 0 0 0 0
echelon forms.
7
 Uniqueness of RREF
Key Points:
➢It is always possible to convert an 𝑚 × 𝑛 matrix to row echelon or reduced row
echelon form.
➢If a matrix is in reduced echelon form then it must be in echelon form.
➢The reduced row echelon form of a matrix is unique.
➢The row echelon form of a matrix is not unique.
➢REF/ RREF of a matrix 𝐴 is also row equivalent to 𝐴.
➢The leading entry is also called a pivot.
Row Equivalent:
Let 𝐴 be an 𝑚 × 𝑛 matrix. An 𝑚 × 𝑛 matrix 𝐵 is called row equivalent to 𝐴 if 𝐵 is
obtained from 𝐴 by performing a finite sequence of elementary row operations on 𝐴.
𝑅
We write 𝐵 ~ 𝐴 to denote 𝐵 is row equivalent to 𝐴.

8
 Elementary Row Operations
Definition:
The following operations on a matrix 𝐴 are called elementary row operations:
i. Interchange of any two rows (e.g. 𝑅13 ).
ii. Multiplication of a row of 𝐴 by any nonzero number (e.g. 9𝑅2 ).
iii. Addition of a scalar multiple of one row to another row (e.g. 𝑅2 +4𝑅1 ).

1 0 0
Elementary Matrices: Consider the identity matrix 𝐼3 = 0 1 0 . The
0 0 1
following matrices 𝐸1 , 𝐸2 , 𝐸3 obtained from 𝐼3 are called elementary matrices of
order 3.

9
 Elementary Matrices
1 0 0
• 𝐸1 = 0 0 1 , Interchange of two rows of 𝐼3 . (Here 𝑅2 and 𝑅3 are
0 1 0
interchanged i.e. 𝑅23 )

1 0 0
• 𝐸2 = 0 3 0 , Multiply a row of 𝐼3 by a nonzero number. (Here 𝑅2 is
0 0 1
multiplied by 3 i.e. 3𝑅2 )

1 0 0
• 𝐸3 = 0 1 0 , Add a multiple of a row of 𝐼3 to another row. (Here −5𝑅2 is
0 −5 1
added to 𝑅3 i.e. 𝑅3 +(−5)𝑅1 ).
10
 Echelon and Reduced Echelon Forms
Question-01:
1 2 −3 1 2
Reduce the matrix; 𝐴 = 2 4 −4 6 10 into echelon form.
3 6 −6 9 13
Solution:
𝟏 2 −3 1 2
• Select the leading entry in 1st row; 2 4 −4 6 10
3 6 −6 9 13
• Now make all the entries zero that are below the leading entry.
Working: 𝑅2 − 2𝑅1
𝟏 2 −3 1 2 𝑅2 − 2𝑅1 2 – 2 1 =0, 4– 2 2 =0
𝟎 0 2 4 6 𝑅 − 3𝑅 –4 –2 –3 = 2 , 6– 2 1 =4
3 1
𝟎 0 3 6 7 10 – 2(2) = 6

11
 Echelon and Reduced Echelon Forms
1 2 −3 1 2
• Select the leading entry in 2nd row; 0 0 𝟐 4 6
0 0 3 6 7
• Now make all the entries zero that are below the leading entry.

1 2 −3 1 2
3 𝟑
0 0 𝟐 4 6 , by 𝑅3 − 𝑅 𝒆. 𝒈. 𝟑 – (𝟐) = 𝟎
2 2 𝟐
0 0 𝟎 0 −2

1 2 −3 1 2
• Now the matrix 𝐴 is in row echelon form i.e. 0 0 2 4 6 .
0 0 0 0 −2
• We can further reduce the above matrix 𝐴 to reduced echelon form.

12
 Echelon and Reduced Echelon Forms
Question-02:
1 2 −3 1 2
Reduce the matrix; 𝐴 = 2 4 −4 6 10 into reduced echelon form.
3 6 −6 9 13
Solution:
𝟏 2 −3 1 2
• Select the leading entry in 1st row; 2 4 −4 6 10
3 6 −6 9 13
• Now make all the entries zero that are below the leading entry.
Working: 𝑅2 − 2𝑅1
𝟏 2 −3 1 2 𝑅2 − 2𝑅1 2 – 2 1 =0, 4– 2 2 =0
𝟎 0 2 4 6 𝑅 − 3𝑅 –4 –2 –3 = 2 , 6– 2 1 =4
3 1
𝟎 0 3 6 7 10 – 2(2) = 6

13
 Echelon and Reduced Echelon Forms

• Select the leading entry in 2nd row and make it 1 (if it is not 1).

1 2 −3 1 2 1 2 −3 1 2
1
0 0 𝟐 4 6 ⟹ 0 0 𝟏 2 3 , 2 𝑅2
0 0 3 6 7 0 0 3 6 7

• Now make all the entries zero that are above and below the leading entry.

1 2 𝟎 7 11
𝑅1 + 3𝑅2
0 0 𝟏 2 3
0 0 𝟎 0 −2 𝑅3 − 3𝑅2

14
 Echelon and Reduced Echelon Forms

• Select the leading entry in 3rd row and make it 1 (if it is not 1).

1 2 0 7 11 1 2 0 7 11 1
0 0 1 2 3 ⟹ 0 0 1 2 3 , − 𝑅3
2
0 0 0 0 −𝟐 0 0 0 0 𝟏

• Now make all the entries zero that are above and below the leading entry.

1 2 0 7 𝟎 𝑅 − 11𝑅
1 3
0 0 1 2 𝟎
𝑅2 − 3𝑅3
0 0 0 0 𝟏

15
 Echelon and Reduced Echelon Forms

1 2 0 7 0
0 0 1 2 0
0 0 0 0 1

• Now the matrix 𝐴 is in row reduced echelon form.

16
 Rank of a Matrix
Definition:
The rank of a matrix 𝐴 is the number of nonzero rows (or columns) in its echelon
(or reduced echelon) form. It is written as rank(𝐴). Rank is very helpful in solving
linear systems.
Key Points:
➢To find the rank of an 𝑚 × 𝑛 matrix A, we just convert A to its echelon (or
reduced echelon) form and count its nonzero rows.
➢The rank of an 𝑚 × 𝑛 matrix is a nonnegative integer and cannot be greater than
either m or n.
➢Only a zero matrix has rank zero.
➢rank(𝐴) = rank (𝐴𝑇 )

17
 Rank of a Matrix
Question-03:
1 3 1 −2 −3
Find rank of the matrix, 𝐴= 1 4 3 −1 −4 .
2 3 −4 −7 −3
3 8 1 −7 −8

Solution:
• Here we find the rank by converting 𝐴 into RREF. Select the leading entry in 1st
row (it must be 1).
𝟏 3 1 −2 −3
1 4 3 −1 −4
2 3 −4 −7 −3
3 8 1 −7 −8

18
 Rank of a Matrix
• Now make all the entries zero that are above & below the leading entry.

𝟏 3 1 −2 −3
𝟎 1 2 1 −1 𝑅2 − 𝑅1
𝟎 −3 −6 −3 3 𝑅3 −2 𝑅1
𝟎 −1 −2 −1 1 𝑅4 − 3 𝑅1

• Next choose the leading entry in 2nd row (it must be 1).

1 3 1 −2 −3
0 𝟏 2 1 −1
0 −3 −6 −3 3
0 −1 −2 −1 1

19
 Rank of a Matrix
• Now make all the entries zero that are above & below the leading entry.

1 𝟎 −5 −5 0
𝑅1 − 3𝑅2
0 𝟏 2 1 −1
0 𝟎 0 0 0 𝑅3 + 3𝑅2
0 𝟎 0 𝑅4 + 𝑅2
0 0
1 0 −5 −5 0
• Now the matrix 𝐴 is in RREF i.e. 0 1 2 1 −1 , also this form has two
0 0 0 0 0
0 0 0 0 0
nonzero rows & two zero rows.
• So, rank (𝐴) = Number of nonzero rows in RREF = 2

20
 Practice Questions
Questions:
Find the rank of the following matrices. 1 3 1 −2 −3
5 9 3 1 4 3 −1 −4 (rank = 2)
1. −3 4.
5 6 (rank = 2) 2 3 −4 −7 −3
−1 −5 −3 3 8 1 −7 −8
1 3
1 2 −3
2. 0 −2 (rank = 2)
5 −1 5. 2 1 0 (rank = 3)
−2 −1 3
−2 3
−1 4 −2
2 4 8 16
3. 16 8 4 2 (rank = 4) 1 −2
4 8 16 2 6. 0 0 (rank = 1)
2 16 8 4 −3 6
21