Linear Algebra (MA1101)

A J Sanasam
anand_sanasam@[Link]
October 10, 2023

Row Operations and System of Linear Equations

Learning outcomes of this material. After studying this mate-

rial (and the questions that follow) thoroughly, the students will be
able to

1. write the conditions for which a given system of linear equations

has (a) a unique solution, (b) infinitely many solutions, and (c)
no solution,

2. solve a system of linear equations by Gauss elimination and back

substitution,

3. find the inverse of a square matrix by using elementary row oper-

ations,

4. solve a system of n linear equations in n variables by using the

inverse of the coefficient matrix, provided the inverse exists.

Read the following notes carefully.

1 System of Linear Equations

A linear equation is an equation of the form

a1 x1 + · · · + an xn = b,

where x1 , . . . , xn are the unknowns or the variables, and a1 , . . . , an , b

are constants. Graphically, such an equation represents a point (one
variable in consideration), a line (two variables in consideration), a
plane (three variables in consideration) or a hyperplane. A solution
of a linear equation is a point or a vector which satisfies the equation,
i.e., a point which lies on the graph of the equation.

Linear Algebra (MA1101) Page 1 of 15

A system of linear equations is a set of linear equations, and a solu-
tion of a system of linear equations is a point (vector) which satisfies
all the equations simultaneously. Systems of linear equations widely
occur in basic sciences and engineering. In this section, we shall
study the importance of matrices in solving systems of linear equa-
tions.
Definition 1.0.1. A system of m linear equations in n unknowns
(variables) x1 , . . . , xn is a set of equations of the form
a11 x1 + · · · + a1n xn = b1 ,
..
. (1)
am1 x1 + · · · + amn xn = bm ,
where aij ’s and bi ’s are constants. If bi = 0 for all i, then (1) is called
a homogeneous system; otherwise, (1) is called a nonhomogeneous
system.
Definition 1.0.2 (Solution of a system of linear equations). A
solution of a system of linear equations is a set of numbers x1 , . . . , xn
or a point (x1 , . . . , xn ), where all the equations are simultaneously
satisfied.
For example,
2x1 − 3x2 + x3 = 0,
3x1 − 2x2 + x3 = 2,
3x1 + x2 + 2x3 = 6
is a system of three linear equations in three variables. Notice that
the numbers x1 = 1, x2 = 1 and x3 = 1 form a solution of the above
system. Note that each of the above linear equations represents a
plane in space and the solution (1, 1, 1) is a point that lies in the
intersection of the three planes represented by the above equations.

1.1 Matrix Form of a System of Linear Equations

The system of linear equations given in definition 1.0.1 can also be
written in matrix form as
Ax = b,

Manipur Technical University Page 2 of 15

where
 
a11 a12 · · · a1n    
x1 b1
a21 a22 · · · a2n ..  , b =  ..  .
 
A = [aij ] =  .. .. .. .. , x =  .  . 
  
. . . .

 
xn bm
am1 am2 · · · amn

The matrix A is called the coefficient matrix. The matrix [A|b]

defined by
 
a11 a12 · · · a1n b1
 a21 a22 · · · a2n b2 
[A|b] = [aij |bi ] =  .. .. .. .. ..
 
 . . . . .


am1 am2 · · · amn bm

is called the augmented matrix of the system Ax = b. The aug-

mented matrix contains all the coefficients, and it completely deter-
mines the system Ax = b.

1.2 Elementary Row Operations

There are three basic elementary row operations which will transform
the given system into another system without changing its solutions.
The elementary row operations are
1. interchanging two rows,
2. multiplication of a row by a non zero constant, and
3. addition of a row by a constant multiple of another row.
Definition 1.2.1. Two systems of linear equations or matrices are
said to be row equivalent if one can be obtained from the other by a
finite number of elementary row operations.
Theorem 1.1. Row equivalent systems of linear equations have the
same set of solutions.
Now, we shall see how one can use matrices in solving a given system
of linear equations.

Linear Algebra (MA1101) Page 3 of 15

1.3 Row Operations and Gauss Elimination
Definition 1.3.1. A matrix is said to be in row echelon form if

(a) all rows consisting of only zeroes are at the bottom,

(b) the leading coefficient (also called the pivot) of a nonzero row is
always strictly to the right of the leading coefficient of the row
above it.
For example, the matrix
 
0 1 1 9 0
 0 0 0 7 8 
0 0 0 0 9

is in row echelon form.

If a system Ax = b of linear equations is such that the coefficient
matrix A is in the row echelon form, then we can find its solution
instantly.

Example 1.1. Consider the following system of linear equations.

x1 + 2x2 + 6x3 = 9,
x2 + x3 = 2,
x3 = 1.

The coefficient matrix is in row echelon form. The solution is given

by
x3 = 1, x2 = 1, x1 = 1.

Theorem 1.2. Every matrix is row equivalent to a matrix in row

echelon form.

1.4 Gauss Elimination and Back Substitution

Consider a system of linear equations

Ax = b. (2)

Manipur Technical University Page 4 of 15

To find a solution of this system, we use elementary row operations
to reduce the augmented matrix [A|b] to a matrix [R|f ] of the form

r12 · · · r1k · · · r1n f1

 
r11
 0
 . r22 · · · r2k · · · r2n f2 
 . .. . . .. .. .. 
 . . . . . . 

[R|f ] =  0 0 · · · rkk · · · rkn fk  ,
 
 0 0 · · · 0 · · · 0 fk+1 
 
 . .. .. .. .. 
 .. . . . . 
0 0 ··· 0 ··· 0 fm

where k called the rank of A and it is the number of non zero rows
of the row echelon form R of A. The reduction of [A|b] to [R|f ]
by a finite number of elementary row operations is known as Gauss
elimination, and the system [R|f ] is called row echelon form of the
given system.
Now, two possibilities arise for the solutions of the system Ax = b.
Inconsistent: The system Ax = b will have no solution if k < m
and at least one of the numbers fk+1 , . . . , fm is non zero. In this case,
the system does not have any solution and we say that the system
Ax = b is inconsistent. See example 1.2, on page 6.
Consistent: If k ≤ m and fk+1 = · · · = fm = 0, then the system
has solution(s) and we say that the system Ax = b is consistent.
Two cases arise.
(a) The system Ax = b has a unique solution if k = n. In this
case, xn = rfnn
n
. Substituting back this value in the (n − 1)th row,
we can find the value of xn−1 . Again, substituting these values back
in the (n − 2)th row, we can find the value of xn−2 . Continuing this
process, we can find all the values of the variables. This process of
substituting the known variables back and finding the values of the
remaining variables is known as back substitution. See example 1.3,
on page 6.
(b) The system Ax = b has infinitely many solutions if k < n. In
this case, n − k variables namely, xk+1 , . . . , xn can be given arbitrary

Linear Algebra (MA1101) Page 5 of 15

values. By back substitution, we can find the values of the remaining
variables in terms of these variables. See example 1.4, on page 7.

Example 1.2. Find the solution of the following system of linear

equations.
x + y + z = 4,
x − y + 3z = 6,
3x + y + 5z = 15.

Solution: The augmented matrix [A|b] of the given system is

 
1 1 1 4
[A|b] =  1 −1 3 6  .
3 1 5 15

We use row operations to reduce this system to an equivalent matrix

which is in the row echelon form as follows.
 
1 1 1 4
[A|b] =  1 −1 3 6 
3 1 5 15
 
1 1 1 4
∼  0 −2 2 2  R2 − R1 , R3 − 3R1
0 −2 2 3
 
1 1 1 4
∼  0 −2 2 2  R3 − R2 .
0 0 0 1

From this row echelon form [R|f ], we notice that k = 2 < 3 = m and
f3 6= 0. Hence, the given system has no solution.

Example 1.3. Find the solution of the following system of linear

equations.
x + 2y + 2z = 4,
y + z = 1,
x + 4y − z = −4,
x + 4y + 5z = 8.

Manipur Technical University Page 6 of 15

Solution: The augmented matrix [A|b] of the given system is
 
1 2 2 4
 0 1 1 1 
[A|b] = 
 1 4 −1 −4  .


1 4 5 8
We use row operations to reduce this system to an equivalent matrix
which is in the row echelon form as follows.
 
1 2 2 4
 0 1 1 1 
[A|b] =  1 4 −1 −4 


1 4 5 8
 
1 2 2 4
 0 1 1 1 
∼  0 2 −3 −8  R3 − R1 , R4 − R1


0 2 3 4
 
1 2 2 4
 0 1 1 1 
∼  0 0 −5 −10  R3 − 2R2 , R4 − 2R2


0 0 1 2
 
1 2 2 4
 0 1 1 1  1
∼  0 0 −5 −10  R4 + 5 R3 .


0 0 0 0
From this row echelon form [R|f ], we see that k = 3 = n and f4 = 0.
So, the given system has a unique solution. By back substitution,
we have −5z = −10, i.e., z = 2, y = 1 − 2 = −1 and x = 4 − 2(−1) −
2(2) = 2. Hence, the solution of the given system is given by
x = 2, y = −1, z = 2.
Example 1.4. Find the solution of the following system of linear
equations.
x + y + z = 4,
x − y + 3z = 6,
3x + y + 5z = 14.

Linear Algebra (MA1101) Page 7 of 15

Solution: The augmented matrix [A|b] of the given system is
 
1 1 1 4
[A|b] =  1 −1 3 6  .
3 1 5 14
We use row operations to reduce this system to an equivalent matrix
which is in the row echelon form as follows.
 
1 1 1 4
[A|b] =  1 −1 3 6 
3 1 5 14
 
1 1 1 4
∼  0 −2 2 2  R2 − R1 , R3 − 3R1
0 −2 2 2
 
1 1 1 4
∼  0 −2 2 2  R3 − R2 .
0 0 0 0
From this row echelon form [R|f ], we see that k = 2 < 3 = m
and f3 = 0. So, the given system has infinitely many solutions
and n − k = 1 variable, say z can be given arbitrary values. By
back substitution, we have −2y + 2z = 2, i.e., y = z − 1, and
x = 4 − z − (z − 1) = 5 − 2z. Hence, the solution of the given system
is given by
x = 5 − 2z, y = z − 1, z = z, where z can take any value.

1.5 Inverse of a Matrix

Definition 1.5.1 (Inverse of a matrix). A square matrix A of
order n is called invertible if there exists a matrix B of order n such
that AB = BA = I, where I is the identity matrix of order n. The
matrix B, if it exists, is uniquely determined and is called inverse of
A, and it is denoted by A−1 .
   
1 2 3 −2
For example, the inverse of A = is A =
−1
1 3 −1 1
because AB = BA = I.

Manipur Technical University Page 8 of 15

Theorem 1.3. The inverse of a matrix A, if it exists, is unique.
Proof: Let B and C be inverses of A. By definition, we have
AB = BA = I, AC = CA = I, where I is the identity matrix.
Now, B = BI = B(AC) = (BA)C = IC = C. Hence, the inverse of
a matrix, if it exists, is unique.

1.6 Elementary Matrices

Consider the following matrices
   
1 2 3 0 1 0
A= 4  5 6  and P =  0 0 0  .
7 8 9 0 0 0
Then
    
0 1 0 1 2 3 4 5 6
PA =  0 0 0  4 5 6  =  0 0 0 .
0 0 0 7 8 9 0 0 0
It may be observed that P is the matrix all of whose entries are zero
except p12 = 1. Again, it may be observed that P A is the matrix
obtained by putting the 2nd row of A in the 1st row, and zeros
elsewhere. What would happen if all the entries of P are zero except
pij = k (a constant)?
Now, let us observe the folloing multiplications of the form P A.
    
1 0 0 1 2 3 1 2 3
PA = 0  0 1   4 5 6 = 7
  8 9 , (3)
0 1 0 7 8 9 4 5 6
    
3 0 0 1 2 3 3 6 9
PA = 0  1 0   4 5 6 = 4
  5 6 , (4)
0 0 1 7 8 9 7 8 9
    
1 2 0 1 2 3 9 12 15
PA =  0 1 0  4 5 6  =  4 5 6 . (5)
0 0 1 7 8 9 7 8 9

Linear Algebra (MA1101) Page 9 of 15

From the first equation, it may be noticed that the matrix P is
obtained from the identity matrix by interchanging the 2nd row and
the 3rd row. What is the effect of P ’s action (left multiplication) on
A?
From the second equation, it may be noticed that the matrix P is
obtained from the identity matrix by multiplying the 1st row by 3.
What is the effect of P ’s action (left multiplication) on A?
From the third equation, it may be noticed that the matrix P is
obtained from the identity matrix by the row operation
addition of the 1st row by 2 times the 2nd row.
What is the effect of P ’s action (left multiplication) on A? Again,
what whould happen if P is a matrix obtained from the identity matrix
by an elementary row operation?
Definition 1.6.1 (Elementary Matrix). An elementary matrix is
a matrix obtained from the identity matrix by an elementary row
operation.
For example, them matrix P in each of the equations (3), (4) and
(5) is an elementary matrix.
Theorem 1.4. Let E be an elementary matrix of order n obtained
from the identity matrix by an elementary row operation, say ∗. Let
A be a square matrix of order n. Then EA is the matrix obtained
from A by applying the elementary row operation ∗.
Thus, we can think of elementary row operations as elementary ma-
trices, and vice-versa.
Theorem 1.5. An elementary matrix is invertible.
Theorem 1.6. Let A1 , . . . , Ak be invertible matrices of the same
size. Then the product A1 · · · Ak is invertible, and

(A1 · · · Ak )−1 = A−1 −1

k · · · A1 .

Manipur Technical University Page 10 of 15

Theorem 1.7. Let A be a square matrix and B be a matrix row
equivalent to A. Then A is invertible if and only if B is invertible.

Theorem 1.8. A square matrix A is invertible if and only if A is

row equivalent to the unit matrix.

1.7 Inverse of a Matrix by Gauss-Jordan Elimi-

nation
Consider an invertible square matrix A of order n. Then the matrix
A is row equivalent to I, the identity matrix of order n. Thus,
there is a sequence of row operations (or elementary matrices), say
E1 , E2 , . . . , Ek such that

E1 E2 · · · Ek A = I.

Now multiplying both sides from the right by A−1 , we have

E1 E2 · · · Ek AA−1 = IA−1 =⇒ E1 E2 · · · Ek I = A−1 .

Thus, if we can reduce the augmented matrix [A|I] to a row equiv-

alent matrix of the form [I|B] for some matrix B, then B is the
inverse of A.

Example 1.5. Let  

1 1 2
A =  2 1 2 .
3 2 1
Find the inverse of A, if it exists, by Gauss Jordan elimination.

Solution: We proceed as follows.

 
1 1 2 1 0 0
[A|I] =  2 1 2 0 1 0 
3 2 1 0 0 1
 
1 1 2 1 0 0
∼  0 −1 −2 −2 1 0  R2 − 2R1 , R3 − 3R1
0 −1 −5 −3 0 1

Linear Algebra (MA1101) Page 11 of 15

  
1 1 2 1 0 0
∼  0 −1 −2 −2 1 0  R3 − R2
0 0 −3 −1 −1 1
 
1 1 2 1 0 0
1
∼  0 1 2 2 −1 0  − R3
3
0 0 1 13 1
3
− 13
 
1 1 0 13 − 23 − 23
∼  0 1 0 43 − 53 − 23  R2 − 2R3 , R1 − 2R3
 
0 0 1 13 1
3
− 13
 
1 0 0 −1 1 0
4
∼ 0 1 0

3
− 53 2 
3  R1 − R2 .
1 1 1
0 0 1 3 3
−3

Thus, the inverse of the matrix A is given by

 
−1 1 0
 4 5 2 
A−1 =  3 − 3 3 .
1 1
3 3
− 13

1.8 Solution of a System of Linear Equations by

Inverse
For a system Ax = b of n linear equations in n variables (i.e., A
is a square matrix of order n), the solution is given by x = A−1 b,
provided A−1 exists.
Theorem 1.9. Let Ax = b be a system of n linear equations in n
variables. If the inverse of A exists, then the system Ax = b has a
unique solution, and the solution is given by
x = A−1 b.
Proof: If A−1 exists, then multiplying the system Ax = b by A−1
from the left side, we have
A−1 (Ax) = A−1 b =⇒ x = A−1 b.

Manipur Technical University Page 12 of 15

Example 1.6. Solve the system Ax = b, where
     
1 1 2 x1 2
A= 2  1 2 , x = x2 , b = 3 
   
3 2 1 x3 5

by using inverse of A.
Solution: By example 1.5 on page 11, the inverse of A is given by
 
−1 1 0
 4 5 2 
A−1 =  3 − 3 3 .
1 1 1
3 3
− 3

Now, the solution is given by

    
−1 1 0
 
x1 2 1
−1  4 −5 2 
x = x2 = A b =  3
  3 3   3 = 1 .
 
1 1 1
x3 3 3
−3 5 0

Exercise 2.1
1. Find at least one non-trivial solution for each one of the following
systems of equations.
3x1 + x2 + x3 = 0,
(a) 3x1 + x2 + x3 = 0. (b)
x1 + x2 + x3 = 0.
2. In each of the following cases, find a row equivalent matrix in the
row echelon form.
   
  1 1 5 1 0 5 3
2 2 4
(a) . (b)  1 3 3 . (c)  1 2 1 2 .
1 3 2
2 4 1 2 4 1 3
   
−1 0 3   1 1 3
 1 0 0  −2 0 0
(d)  . (e) . (f)  1 1 0 .
 
   −1 0 2
4 0 3   2 2 0 
2 2 0
3 0 1 0 0 1

Linear Algebra (MA1101) Page 13 of 15

3. Solve the following system Ax = b of linear equations.
8x2 + 6x3 = −4,
−3x1 + 8x2 = 5,
(a) (b) −2x1 + 4x2 − 6x3 = 18,
8x1 − 12x2 = −11.
x1 + x2 − x3 = 2.
4x1 + 5x2 = 0, x1 − x2 = 0,
(c) (d)
−6x1 + 7x2 = 0. −3x1 + 5x2 = 2.
 
3 −2 1 2 1
(e) [A|b] = 1
 1 −1 −1 −2 .
2 −1 3 0 4
 
5 −7 3 17
(f) [A|b] = .
−15 21 −9 50
 
2 3 1 −11 1
 5 −2 5 −4 5 
(g) [A|b] = 
 1 −1
.
3 −3 3 
3 4 −7 2 −7
 
10 4 −2 0 14
 −15 1 2 −3 0 
(h) [A|b] =  .
 1 1 0 1 6 
−5 5 −10 8 26
4. For what values of c does the following system have a solution?
x1 − x2 − 3x3 = 3,
2x1 + x3 = 0,
2x2 + 7x3 = c.

5. Find the inverse of each of the following matrices by Gauss-Jordan

elimination.
   
1 4 1 0
(a) . (b) .
2 1 5 3
 
1 0 2  
0 1
(c)  0 1 2 . (d) .
5 0
0 0 1

Manipur Technical University Page 14 of 15

6. Find the solution of Ax = b if
   
1 0 2 2
A= 0 1 2 , b= 3 
  
1 0 1 4

by using inverse of A.

Linear Algebra (MA1101) Page 15 of 15