Module 2: Matrices and Elementary Row Operations

Introduction

Converting a linear system to an equivalent triangular system provides an

algorithm for solving a linear system. This algorithm can be streamlined by
introducing the language of matrices to represent a linear system.

Learning Objectives

At the end of this topic, you should be able to demonstrate the following:

a. Write the augmented matrix of a given linear systems;

b. Identify if a matrix is reduced row echelon form;
c. Transform the augmented matrix in reduced row echelon form; and
d. Solve a linear system using Gauss – Jordan elimination method.

Presentation of Contents

Matrix
Definition. For m and n ≥ 1, an m × n matrix (read as “ m by n matrix”) is a
rectangular array of real numbers with m rows and n columns denoted as

a11 a12 a13 … a1n

a21 a22 a23 … a2n
a31 a32 a33 … a3n
A = ⋮ ⋮ ⋮ ⋮
am1 am2 am3 … amn

Where: A = the name of the matrix (matrices are named using capital
letters)
We number the rows as 1, 2, . . . , m from top to bottom, and the
columns 1, 2, . . . , n from left to right. We refer to the entry in row i and
column j as the (i, j) entry. If the matrix is A, then we record the (i, j) entry
as aij , and often write the matrix as A = [aij].
An m x n matrix A is usually denoted as mAn or Am x n
An n x n matrix A is denoted as An or An x n (the number of rows is
the same with the number of columns). A is called a square matrix.

Note. There are variations on this notation, depending on context and on

personal preference. The same is true for other pieces of mathematical
notation too!
Example:
1 2 3
A3 = 5 1 −1 ; A is a 3 x 3 matrix
0 2 −4

1 2 8
2 B3 = 3 −5 4 ; B is a 2 x 3 matrix

2
3 C1 = −1 ; C is a 3 x 1 matrix
3

D2 = [ 12 01] ; D is a 2 x 2 matrix

Now, we can write the linear system of m equations of n unknown variables

a11 x1 + a12 x2 + a13 x3 + … + a1n xn = b1
a21 x1 + a22 x2 + a23 x3 + … + a2n xn = b2
a31 x1 + a32 x2 + a33 x3 + … + a3n xn = b3
⋮ ⋮ ⋮ ⋮ ⋮
am1 x1 + am2 x2 + am3 x3 + … + amn xn = bm

in matrix form as

a11 a12 a13 … a1n x1 b1

a21 a22 a23 … a2n x2 b2
A= a31 a32 a33 … a3n X= x3 B= b3
⋮ ⋮ ⋮ ⋮ ⋮ ⋮

am1 am2 am3 … amn xm bm

We can write the linear system using the relation AX = B. Where A is the
coefficient matrix, X is the matrix of variables and B is the matrix of constants.
Most often we will dispense with the matrix of variables X and will simply write
the augmented matrix of the system as

a11 a12 a13 … a1n b1

a21 a22 a23 … a2n b2
 a31 a32 a33 … a3n b3
A = ⋮ ⋮ ⋮ ⋮ ⋮
am1 am2 am3 … amn bm

For example, the augmented matrix of the linear system

x1 - 2x2 + x3 = -1
1 −2 1 Note:
−1
2x1 - 3x2 - x3 = 3 is 2 −3 −1 Write
3 0 for missing terms.
1 −2 2 1
x1 - 2x2 + 2x3 = 1
When solving linear system by Gaussian elimination, only the coefficients of the
variables and the constants are considered to find the solution. The variables are
placeholders. We can simplify this by introducing the language of matrix using the
augmented matrix of the linear system. This is because the entries of the augmented
matrix of the system are the coefficients of the variables and the constants.
The method of elimination on a linear system is the same as performing similar
operations on the rows of the corresponding augmented matrix. An illustration is
presented below. Observe the difference and be able to appreciate why augmented matrix
of the system is simpler and easier to work with.

Linear System Corresponding Augmented Matrix

x+y–z= 1 1 1 −1 1
2x – y + z = -1 2 −1 1 −1
−1 −1 3 3
-x – y + 3z = 3

Using the operations: Using the operations:

-2E1 + E2 →E2 -2R1 + R2 →R2
-2x – 2y + 2z = -2 R1 + R3 →R3
2x – y + z = -1
-3y + 3z = -3 We obtain the equivalent augmented
matrix
E1 + E3 →E3

x+y–z= 1
-x – y + 3z = 3
2z = 4

We obtain the triangular system

x+y–z= 1 1 1 −1 1
–3y + 3z = -3 0 −3 3 −3
0 0 2 4
2z = 4

Furthermore, using the operations: Furthermore, using the operations:

 −1 −1
E2 →E2 R2 →R2
3 3
– 3 y +3 z −3 1
= R3 →R3
3 3 2
-y + z = -1
y–z=1 We obtain a more simplified equivalent
augmented matrix
1
E3 →E3
2

We now have a more simplified

equivalent triangular system

x+y–z= 1 1 1 −1 1
y–z=1 0 1 −1 1
0 0 1 2
z=2

Similar to the triangular form of a linear system, a matrix is in triangular form

provided that the first nonzero entry for each row of the matrix is to the right of the first
nonzero entry in the row above it.

Note:

When working with the augmented matrix, the notation E used for linear
systems is changed to R.

The next theorem is a restatement of theorem 1 of module 1 in terms of operations

on the rows of an augmented matrix.

Theorem 2
Any of the following operations performed on the augmented matrix of a linear
system, produces an augmented matrix of an equivalent linear system. These operations
are called elementary row operations.
1. Interchange any two rows.
The notation Ri ↔Rj will be used to indicate that row i and row j can
interchanged.
2. Multiplying any row by a none zero constant.
 The notation cRi, where c ≠0 means multiplying row i by none zero constant
c.
3. Adding a multiple of one row to another.
Example of this is -2R1 + R3 →R3 means add -2 times row 1 to row3 and
replace row 3.

Row Echelon Form of a Matrix

The resulting matrix from the table above after performing several elementary
operations is said to be in row echelon form.

1 1 −1 1
0 1 −1 1
0 0 1 2

Transforming an augmented matrix of linear system in row echelon form still

requires back substitution in order to find the solution of the system.
Now, let us see how elementary row operations can be further use to produce an
equivalent augmented matrix so that we will no longer use back substitution to find the
solution of the system.
 Let use the row echelon form as the starting point to illustrate further the example
above.
Linear System Corresponding Augmented Matrix
Triangular form of the linear system Row echelon form of the augmented
matrix
x+y–z= 1
y–z=1 1 1 −1 1
z=2 0 1 −1 1
0 0 1 2
Performing the operations: Performing the operations:
E1 + (-1)E2 →E1 R1 + (-1)R2 →R1
We obtain the equivalent linear system We obtain an equivalent augmented
matrix
x =0
y–z=1 1 0 0 0
z=2 0 1 −1 1
0 0 1 2
Using the operation: Using the operation:
E2 + E3 →E2 R2 + R3 →R2
We obtain the equivalent linear system We obtain an equivalent augmented
matrix
x =0
y =3 1 0 00
z=2 0 1 03
0 0 12
Using the information from the last row of the table, the solution can now read
directly as x = 0, y = 3, and z = 2 without using back substitution.
The resulting augmented matrix
1 0 00
0 1 03
0 0 12

is called the reduced row echelon form of the matrix. The leading nonzero entry in
each row is called a pivot element. The matrix is in reduced row echelon form if each
pivot is 1 and all other entries on the columns containing the pivot element are zeros.
The process of transforming a matrix to reduced row echelon for is called Gauss-
Jordan elimination.
Here are other examples of matrices in reduced row echelon form.
Reduced Row Echelon of a Matrix
An m x n matrix is in reduced row echelon form if
1. Every row consisting entirely of zeros is found below the rows with nonzero
entries.
2. The leading nonzero entry of each row is 1 called the pivot element.
3. All other entries in the column containing the leading entry 1 are zeros.
4. The leading entry in a row occurs to the right of the leading entry of the row
above it.

Example 1. Given the augmented matrix, find the solution of the corresponding
linear system.
1 0 0 1 1 0 2 −3 1 0 0 −1
a. 0 1 0 2 b. 0 1 −1 2 c. 0 1 0 −3
0 0 1 3 0 0 0 0 0 0 0 1

Solution
a. Reading directly from the augmented matrix, x3 = 3, x2 = 2, x1 = 1. The
solution is (1, 2, 3). The system is consistent and has a unique solution.

1 0 2 −3
b. 0 1 −1 2
0 0 0 0

x + 2z = -3
x = -3 – 2z

y–z=2
y=2+z

(-3- 2z, 2 + z, z)
 b. In this case the system is consistent with infinitely many solutions
because in the last row, the coefficient matrix has rows of zeros and the
corresponding augmented term is zero.
The solution is x2 – x3 = 2, so x2 = 2 + x3 and x1 = -3 – 2x3.The variable x3
is free and the general solution is (-3 – 2x3, 2 + x3, x3) where x3 is any real
number.

1 0 0 −1
c. 0 1 0 −3
0 0 0 1

c. The third row of the matrix corresponds to 0 ≠ 1, which is an inequality.

So, the linear system has no solution.

Example 2. Solve the linear system by transforming the augmented matrix to

reduced row echelon form.
x1 – x2 – 2x3 + x4 = 0
2x1 – x2 – 3x3 + 2x4 = -6
-x1 + 2x2 + x3 + 3x4 = 2
x1 + x2 – x3 + 2x4 = 1

The augmented matrix of the linear system is

1 -1 -2 1 0
2 -1 -3 2 -6
-1 2 1 3 2
1 1 -1 2 1

To transform the matrix into reduced row echelon form, we first use the
leading entry in row 1 as a pivot and eliminate the terms in column 1 of rows
2, 3 and 4. To do this we used the operations

x1 – x2 – 2x3 + x4 = 0
2x1 – x2 – 3x3 + 2x4 = -6
 -x1 + 2x2 + x3 + 3x4 = 2
x1 + x2 – x3 + 2x4 = 1

-2R1 + R2 → R2
2x1 – 2x2 – 4x3 + 2x4 = 0
2x1 – x2 – 3x3 + 2x4 = -6
(-x2 -x3 = 6) -1 x2 + x3 = -6

R1 + R3 → R3
x1 – x2 – 2x3 + x4 = 0
-x1 + 2x2 + x3 + 3x4 = 2
x2 –x3 + 4x4 = 2

-R1 + R4 → R4

x1 – x2 – 2x3 + x4 = 0
x1 + x2 – x3 + 2x4 = 1
-2x2 -x3 – x4 = -1
2x2 + x3 + x4 = 1
The equivalent augmented matrix is

1 -1 -2 1 0
0 1 1 0 -6
0 1 -1 4 2
0 2 1 1 1

For the second step we use the leftmost 1 in row 2 as the pivot and eliminate
the term in column 2 above the pivot and the two terms below the pivot. The
operations are

R1 + R2 → R1
X1 –x2 -2x3 + x4 = 0
X2 + x3 =-6
X1 - x3 + x4 = -6

-R2 + R3 → R3
-X2 - x3 = 6
x2 –x3 + 4x4 = 2
-2x3 + 4x4 = 8
-2R2 + R4 → R4

-2X2 -2x3 = 12
2x2 + x3 + x4 = 1
-x3 + x4 = 13
 Again, the equivalent augmented matrix is

1 0 -1 1 -6
0 1 1 0 -6
0 0 -2 4 8
0 0 -1 1 13

Notice that each entry in row 3 is even, therefore to make the leading entry
be 1, use the operation
−1
R3 → R3
2

The resulting matrix is

1 0 -1 1 -6
0 1 1 0 -6
0 0 1 -2 -4
0 0 -1 1 13

Now, by using the leading entry 1 in row 3 as the pivot, perform the
following operations to eliminate the two entries in column 3 above the
pivot and the term below the pivot in row 4.

R1 + R3 → R1
x1 – x3 + x4 = -6
x3 -2x4 = -4
x1 - x4 = - 10

-R3 + R2 → R2
-X3 + 2x4 = 4
X2 - x3 = -6
X2 + 2x4 = -2
R3 + R4 → R4 -x4 = 9

To reduce the matrix is

1 0 0 -1 -10
0 1 0 2 -2
0 0 1 -2 -4
0 0 0 -1 9

Using the operation - R4 → R4, we make the leading entry in row 4 as 1,

the matrix obtain is in row echelon form.
 1 0 0 -1 -10
0 1 0 2 -2
0 0 1 -2 -4
0 0 0 1 -9

Finally, using the leading entry in row 4 as the pivot, we eliminate the terms
above it in column 4. Performing the operations below
R1 + R4 → R1
-2R4 + R2 → R2
2R4 + R3 → R3

gives the augmented matrix in reduced row echelon form

1 0 0 0 -19
0 1 0 0 16
0 0 1 0 -22
0 0 0 1 -9

The solution can now be read directly from the reduced row echelon form.
x4 = -9, x3 = -22, x2 = 16 and x1 = -19. So, the linear system is consistent and
with a unique solution.

Example 3. Solve the linear system

x+ y+ z =4
3x – y – z = 2
x + 3y + 3z = 8

The augmented matrix is

1 1 1 4
3 -1 -1 2
1 3 3 8

Solution
To solve the system, first, we reduce the augmented matrix to row echelon
form. The following steps will illustrate the process.
-3R1 + R2→ R2
-3x1 + -3x2 + -3x3 = -12
3x1 - x2 - x3 = 2
-4x2 -4x3 = -10

- R1 + R3 → R3
-x1 – x2 – x3 = -4
X1 + 3x2 + 3x3 = 8
2x2 + 2x3 = 4
 1 1 1 4
0 -4 -4 -10
0 2 2 4

R2 ↔ R3

1 1 1 4
0 2 2 4
0 -4 -4 -10
1
R → R2
2 2

1 1 1 4
0 1 1 2
0 -4 -4 -10

4R2 + R3 → R3
4x2 + 4x3 = 8
-4x2 – 4x3 = 10

1 1 1 4
0 1 1 2
0 0 0 -2

The third row of the last matrix corresponds to 0 ≠-2. The system therefore
is inconsistent and has no solution. If this case happens, do not continue
anymore transforming the matrix to reduced row echelon form.

Fact Summary

1. Performing the elementary row operations to the augmented matrix will not change
the solutions of the corresponding linear system.

2. Every matrix has a unique reduced row echelon form.

3. If the augmented matrix in reduced row echelon form, the coefficient matrix has no
rows of zeros, the linear system has a unique solution.

4. If the augmented matrix in reduced row echelon form, the coefficient matrix has a
row of zeros and the corresponding augmented term is not zero, the linear system has
no solution.

5. If the augmented matrix in reduced row echelon form, the coefficient matrix has a
 row of zeros and the corresponding augmented term is zero, the linear system has
infinitely many solutions.

Application

Activity 1. Write the augmented matrix of the given linear system. Do not solve the
system. Write your answer on the indicated column in the table below.

Linear System Augmented Matrix

1.
2x – 3y = 5
-x + y = -3
2.
2x – z = 4
x + 4y + z = 2
4x + y – z = 1
3.
3x1 – 3x3 + 4x4 = -3
-4x1 + 2x2 – 2x3 – 4x4 = 4
4x2 – 3x3 + 2x4 = -3

Activity 2. Identify if the matrix is in reduced row echelon form. Put a check ( / ) mark if
the matrix is in reduced row echelon form. Write a cross mark (x) if the matrix in not in
reduced row echelon form and give one reason why it is not

1 2 0
1. 0 0 2 _______
0 0 0

1 2 0
2. 0 0 1
________

3.
1 0 5 0 -1
0 1 -3 0 1 _______
0 0 0 1 -4
 4.
1 0 0 4 -1
0 0 1 5 2 _______
0 1 0 0 -1

5.
1 0 2
0 1 3

Activity 3. Write the linear system as an augmented matrix. Use Gauss-Jordan

elimination to solve the linear system. Describe if the linear system has a unique solution,
no solution or has infinitely many solutions.

1. x+ y =1
4x + 3y = 2

2.
3x – 3y = 3
4x – y – 3z = 3
-2x – 2y = -2

3. x + 2y + z = 1
2x + 3y + 2z = 0
x+y+z=2

4.
-x1 + 3x3 + x4 = 2
2x1 + 3x2 – 3x3 + x4 = 2
2x1 – 2x2 – 2x3 – x4 = -2