Lecture Notes on

Linear Algebra
Theory, Applications, and Complementarity Problems

Prepared by:

Dr. Rashid A
Assistant Professor,
Department of Computer Science and Business Systems
KIT’s College of Engineering Kolhapur (Empowered Autonomous)

Prepared for:

Semester 2, B.Tech. in Computer Science and Business Systems,

Department of Computer Science and Business Systems
KIT’s College of Engineering Kolhapur (Empowered Autonomous)

”Mathematics is not about numbers, equations, computations,

or algorithms: it is about understanding.”
— William Paul Thurston
 REFERENCES

1. Larson, R., Edwards, B. H., and Falvo, D. C. (2016). Elementary Linear Algebra
(8th ed.). Houghton Mi!in.

2. Strang, G. (2023). Introduction to Linear Algebra (6th ed.). Wellesley-Cambridge

Press.

3. Ho”man, K., and Kunze, R. (1971). Linear Algebra (2nd ed.). Pearson.

4. Friedberg, S. H., Insel, A. J., and Spence, L. E. (2019). Linear Algebra (5th ed.).
Pearson.
Contents

1 THEORY OF MATRICES 1

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Matrices and Matrix operations . . . . . . . . . . . . . . . . . . . . . . . . 1

1.3 Determinant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1.4 Inverse of a matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

1.5 Diagonal and triangular matrices . . . . . . . . . . . . . . . . . . . . . . . 21

2 SYSTEM OF LINEAR EQUATIONS 27

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.2 Introduction to system of linear equations . . . . . . . . . . . . . . . . . . 28

2.3 Linear system in two and three unknowns . . . . . . . . . . . . . . . . . . 30

2.4 Non-homogeneous system . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

2.5 Gauss elimination method for solving a system of linear equations . . . . . 39

2.6 Gauss Jordan elimination method for solving a system of linear equations . 42

2.7 Homogeneous system of linear equations . . . . . . . . . . . . . . . . . . . 46

2.8 More on linear systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

2.9 LU Decomposition method. . . . . . . . . . . . . . . . . . . . . . . . . . . 66

2.10 Rank of a Matrix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
 CHAPTER 1

THEORY OF MATRICES

1.1 Introduction

In linear algebra, matrices play a fundamental role in various branches of math-

ematics, engineering, and computer science, serving as essential tools for solving

systems of equations, transforming geometric objects, and representing complex

data structures. This chapter begins with an introduction to matrices and ma-

trix operations, covering fundamental concepts such as addition, multiplication,

and scalar transformations. The study then progresses to the determinant, a

crucial scalar value associated with a square matrix that provides insight into

properties like invertibility and linear dependence. Next, the inverse of a matrix

is explored, highlighting its significance in solving linear systems and transform-

ing vector spaces. Finally, the chapter concludes with a discussion on diagonal

and triangular matrices, which simplify computations and play a key role in

matrix factorization techniques. These topics form the foundation for advanced

studies in linear algebra and its diverse applications.

1.2 Matrices and Matrix operations

matrix is a rectangular array of numbers. The numbers in the array are called

the entries in the matrix. The size of a matrix is described in terms of the

number of rows (horizontal lines) and columns (vertical lines) it contains.

1
2
2 3 2 3
p
6 1 27  6e ⇡ 27 2 3 
6 7 6 7 617
6 3 07 2 1 0 60 1/2 1 7 are
6 7 3 6 7 4 5 4
4 5 4 5 3
1 4 0 0 0
some examples.
The first matrix in example has three rows and two columns, so
its size is 3 by 2 (written 3 ⇥ 2). In a size description, the first
number always denotes the number of rows, and the second
denotes the number of columns. The remaining matrices in
example have sizes 1 ⇥ 4, 3 ⇥ 3, 2 ⇥ 1, and 1 ⇥ 1, respectively.
A matrix with only one row, such as the second in Example, is
called a row vector (or a row matrix), and a matrix with only
one column, such as the fourth in that example, is called a
column vector (or a column matrix). The fifth matrix in that
example is both a row vector and a column vector.
In general,
2 a matrix with3m rows and n columns is written as

6 a11 a12 . . . a1n 7

6 7
6a 7
6 21 a22 . . . a2n 7
A=6 7.
6... ... ... ...7
6 7
4 5
am1 am2 . . . amn
A matrix A with n rows and n columns is called a square
matrix of order n, and the entries a11 , a22 , ..., ann in are said to
 3

be on the main diagonal of A.

Two matrices are defined to be equal if they have the same
size and their corresponding entries are equal.
If A and B are matrices of the same size, then the sum A
+ B is the matrix obtained by adding the entries of B to
the corresponding entries of A, and the di↵erence A - B is
the matrix obtained by subtracting the entries of B from the
corresponding entries of A. Matrices of di↵erent sizes cannot
be added or subtracted.
If A is any matrix and c is any scalar, then the product cA is
the matrix obtained by multiplying each entry of the matrix A
by c. The matrix cA is said to be a scalar multiple of A.
If A is an m ⇥ r matrix and B is an r ⇥ n matrix, then the
product AB is the m ⇥ n matrix whose entries are determined
as follows: To find the entry in row i and column j of AB,
single out row i from the matrix A and column j from the
matrix B. Multiply the corresponding entries from the row and
column together, and then add up the resulting products.

If A1 , A2 , . . . , Ar are matrices of the same size, and if

c1 , c2 , . . . , cr are scalars, then an expression of the form
4

c1 A1 + c2 A2 + . . . + cr Ar is called a linear combination of

A1 , A2 , . . . , Ar with coefficients c1 , c2 , . . . , cr .
If A is an m ⇥ n matrix, and if x is an n ⇥ 1 column vector,
then the product Ax can be expressed as a linear combination
of the column vectors of A in which the coefficients are the
entries of x.
The matrix product can be written as the following linear
combination
2 3of2column
3 2 vectors,
3 for example:
6 1 3 2 76 2 7 6 1 7
6 76 7 6 7
61 2 37 6 7 6 7
6 7 6 17 = 6 97 can be written as
4 54 5 4 5
2 1 2 3 3
2 3 2 3 2 3 2 3
6 17 637 627 617
6 7 6 7 6 7 6 7
2 6 1 7 1 627 + 3 6
6 7 6 7 7 6 7
6 37 = 6 97.
4 5 4 5 4 5 4 5
2 1 2 3
Matrix multiplication has an important application to
systems of linear equations. Consider a system of m linear
equations in n unknowns:
a11 x1 + a12 x2 + . . . + a1n xn = b1
a21 x1 + a22 x2 + . . . + a2n xn = b2
... ... ... ...
am1 x1 + am2 x2 + . . . + amn xn = bm .
 5

The above system of equations can be expressed as a single

matrix
2 equation AX = B3 (3), where
2 3 2 3
6 a11 a12 . . . a1n 7 6 x1 7 6 b1 7
6 7 6 7 6 7
6a 7 6 7 6b 7
6 21 a22 . . . a2n 7 6 x2 7 6 27
A=6 7, X = 6 . 7, B = 6 . 7
6... ... ... ...7 6 .. 7 6 .. 7
6 7 6 7 6 7
4 5 4 5 4 5
am1 am2 . . . amn xn bm
. The matrix A in this equation is called the coefficient
matrix of the system, the matrix X in this equation is called
the variable matrix of the system, the matrix B in this
equation is called the constant matrix of the system.
If A is any m ⇥ n matrix, then the transpose of A, denoted
by AT , is defined to be the n ⇥ m matrix that results by
interchanging the rows and columns of A; that is, the first
column of AT is the first row of A, the second column of AT is
the second row of A, 2and so3 forth.
2 3
62 37
6 7 6 2 1 57
For example: If A = 661 47
7 , then AT
= 4 5
4 5 3 4 6
5 6

If A is a square matrix, then the trace of A, denoted by

tr(A), is defined to be the sum of the entries on the main
6

diagonal of A. The trace of A is undefined if A is not a square

matrix. 2 3 2 3
6a11 a12 a13 7 6 1 3 27
6 7 6 7
Examples: Let A = 6 6a21 a22 a23 7
7 , B = 6
61 2 377,
4 5 4 5
a31 a32 a33 2 1 2
then tr(A)=a11 + a22 + a33 and tr(B)=(-1)+2+(-2)=-1.

Determine whether the statements given below are true or false.

2 3
61 2 37
(a) The matrix 4 5 has no main diagonal.
4 5 6
(b) An m ⇥ n matrix has m column vectors and n row vectors.
(c) If A and B are 2 ⇥ 2 matrices, then AB = BA.
(d) The ith row vector of a matrix product AB can be
computed by multiplying A by the ith row vector of B.
(e) For every matrix A, it is true that (AT )T = A.
(f ) If A and B are square matrices of the same order, then
tr(AB) = tr(A)tr(B) (g) If A and B are square matrices of the
same order, then (AB)T = AT BT .
(h) For every square matrix A, it is true that tr(AT ) = tr(A).
(i) If A is a 6 ⇥ 4 matrix and B is an m ⇥ n matrix such that
BT AT is a 2 ⇥ 6 matrix, then m = 4 and n = 2.
 7

(j) If A is an n ⇥ n matrix and c is a scalar,

then tr(cA) = c tr(A).
(k) If A, B, and C are matrices of the same size such that
A - C = B - C, then A = B.
(l) If A, B, and C are square matrices of the same order such
that AC = BC, then A = B. (m) If AB + BA is defined, then
A and B are square matrices of the same size.
(n) If B has a column of zeros, then so does AB if this product
is defined.
(o) If B has a column of zeros, then so does BA if this product
is defined.
Answers: (a) True (b) False (c) False (d) False (e) True (f)
False (g) False (h) True (i) True (j) True (k) True (l) False
(m)True (n) True (o) False

Algebraic Properties of Matrices

Properties of Matrix Arithmetic
Assuming that the sizes of the matrices are such that the
indicated operations can be performed, the following rules of
matrix arithmetic are valid.
8

(a) A + B = B + A [Commutative law for matrix addition]

(b) A + (B + C) = (A + B) + C [Associative law for matrix
addition]
(c) A(BC) = (AB)C [Associative law for matrix multiplication]
(d ) A(B + C) = AB + AC [Left distributive law]
(e) (B + C)A = BA + CA [Right distributive law]
(f) A(B - C) = AB - AC
(g) (B - C)A = BA - CA
(h) a(B + C) = aB + aC
(i) a(B - C) = aB - aC
(j) (a + b)C = aC + bC
(k) (a - b)C = aC - bC
(l) a(bC) = (ab)C
(m) a(BC) = (aB)C = B(aC)
Commutative law for multiplication will not hold always. The
equality of AB and BA can fail for three possible reasons:
1. AB may be defined and BA may not (for example, if A is 2
⇥ 3 and B is 3 ⇥ 4).
2. AB and BA may both be defined, but they may have
di↵erent sizes (for example, if A is 2 ⇥ 3 and B is 3 ⇥ 2).
 9

3. AB and BA may both be defined and have the same size,

but the two products may be di↵erent.

A matrix whose entries are all zero is called a zero matrix.

Properties of Zero Matrices
If c is a scalar, and if the sizes of the matrices are such that
the operations can be perfomed, then:
(a) A + 0 = 0 + A = A
(b) A - 0 = A
(c) A - A = A + (-A) = 0
(d) 0A = 0
(e) If cA = 0, then c = 0 or A = 0.
Since we know that the commutative law of real arithmetic is
not valid in matrix arithmetic, it should not be surprising that
there are other rules that fail as well. For example, consider
the following two laws of real arithmetic:
If ab = ac and a 6= 0, then b = c. [The cancellation law]
If ab = 0, then at least one of the factors on the left is 0.
The next two examples show that these laws are not true in
matrix arithmetic.
10

Consider
2 the 3 matrices
2 3 2 3
60 1 7 61 1 7 62 57
A=4 5, B = 4 5, C = 4 5.
0 2 3 4 3 4
2 3
63 4 7
Here, AB = AC = 4 5.
6 8
Although A 6= 0, canceling A from both sides of the equation
AB = AC would lead to the incorrect conclusion that B = C.
Thus, the cancellation law does not hold, in general, for matrix
multiplication (though there may be particular cases where it
is true).
Consider the example of two matricesA and B for which
AB =
2 0, 3 A 6= 02and B
but 3 6= 0:
60 17 63 7 7
A=4 5, B = 4 5.
0 2 0 0

A square matrix with 1’s on the main diagonal and zeros

elsewhere is called an identity matrix. An identity matrix is
denoted by the letter
2 I. 3
6 1 0 07
6 7
For example: A = 60 1 07
6
7.
4 5
0 0 1
If it is important to emphasize the size, we will write In for the
 11

n ⇥ n identity matrix.
If A is any m ⇥ n matrix, then AIn = A and Im A = A.

1.3 Determinant

For every square matrix A = [aij ], we associate a number called

determinant of A, denoted by |A| or det(A).
If A = 2
[a11 ], det(A)=a
3 11 .

6a11 a12 7
If A = 4 5, det(A) = a11 a22 a21 a12 .
a21 a22
2 3
6a11 a12 a13 7
6 7
If A = 6
6a21 a22 a23 77, then
4 5
a31 a32 a33

a11 a12 a13

det(A) = a21 a22 a23

a31 a32 a33

a22 a23 a12 a13 a12 a13

= a11 a21 + a31 .
a32 a33 a32 a33 a22 a23

If A is a square matrix, then the minor of entry aij is

12

denoted by Mij and is defined to be the determinant of the

submatrix that remains after the ith row and j th column are
deleted from A. The number (-1)i+j Mij is denoted by Cij and
is called the cofactor of entry aij . 2 3
63 1 47
6 7
Example. Consider the matrix A = 6
62 5 677.
4 5
1 4 8

5 6
The minor of a11 = M11 = = 16.
4 8
The cofactor of a11 = C11 = ( 1)1+1 M11 = M11 = 16.
2 6
The minor of a12 = M12 = = 10.
1 8
The cofactor of a12 = C12 = ( 1)1+2 M12 = M12 = 10.
Similarly, we can find for all other entries of A.

If A is an n ⇥ n matrix, then the number obtained by

multiplying the entries in any row or column of A by the
corresponding cofactors and adding the resulting products is
called the determinant of A, and the sums themselves are
called cofactor expansions of A. That is,
det(A) = a1j C1j + a2j C2j + . . . + anj Cnj , cofactor expansion
 13

along the jth column

det(A) = ai1 Ci1 + ai2 Ci2 + . . . + ain Cin , cofactor expansion
along the ith row.
Example.
2 Find the determinant of the matrix A =
3
63 1 07
6 7
6 2 4 37
6 7
4 5
5 4 2
by cofactor expansion along the first row.
3 1 0
4 3
Solution. det(A) = 2 4 3 = 3
4 2
5 4 2

2 3 2 4
1 +0
5 2 5 4
= 3( 4) (1)( 11) + 0 = 1.

Theorem 1.1. Let A be a square matrix. If A has a row

of zeros or a column of zeros, then det(A) = 0.
Proof. Since the determinant of A can be found by a cofactor
expansion along any row or column, we can use the row or
column of zeros. Thus, if we let C1 , C2 ,...,Cn denote the
cofactors of A along that row or column, then it follows from
14

cofactor expansion that det(A) = 0 C1 + 0 C2 + + 0 Cn

= 0.
Theorem 1.2. Let A be a square matrix. Then det(A) =
det(AT ).
Proof. Since transposing a matrix changes its columns to rows
and its rows to columns, the cofactor expansion of A along
any row is the same as the cofactor expansion of AT along the
corresponding column. Thus, both have the same determinant.
Theorem 1.3. Let A be an n ⇥ n matrix.
(a) If B is the matrix that results when a single row or single
column of A is multiplied by a scalar k, then
det(B) = k det(A).
(b) If B is the matrix that results when two rows or two
columns of A are interchanged, then det(B) = - det(A).
(c) If B is the matrix that results when a multiple of one row
of A is added to another or when a multiple of one column is
added to another, then det(B) = det(A).
Theorem 1.4. If A is a square matrix with two proportional
rows or two proportional columns, then det(A) = 0.
Example matrices having two proportional rows or columns;
 15

thus, each has

2 a determinant
3 of zero.
2 3
61 2 77
6 1 47 6 7
4 5 , 6 4 8 57
6
7
2 8 4 5
2 4 3

1.4 Inverse of a matrix

A square matrix whose determinant is equal to zero is called a

singular matrix.
A square matrix whose determinant is not equal to zero is
called a non-singular
2 3 matrix.
2 3
65 107 63 4 7
Consider A = 4 5 , B = 4 5. Now, det(A) or |A| = 0
3 6 5 8
and det(B) or |B| =
6 0. So A is singular and B is non-singular.

If A is a square matrix, and if a matrix B of the same size

can be found such that AB = BA = I , then A is said to be
invertible (or nonsingular) and B is called an inverse of
A. If no such matrix B can be found, then A is said to be
singular.
The relationship AB = BA = I is not changed by interchanging
16

A and B, so if A is invertible and B is an inverse of A, then it

is also true that B is invertible, and A is an inverse of B. Thus,
when AB = BA = I we say that A and B are inverses of one
another. 2 3 2 3
62 57 63 5 7
Consider the matrices A = 4 5 , B = 4 5. A and B
1 3 1 2
are inverses of each other. One can check it by AB = BA = I.
Theorem 1.5 If B and C are both inverses of the matrix A,
then B = C.
Proof. Since B is an inverse of A, we have BA = I. Multiplying
both sides on the right by C gives (BA)C = IC = C. But it is
also true that (BA)C = B(AC) = BI = B, so C = B.
Remark 1: If A is invertible, then its inverse will be denoted
by the symbol A 1 . Then, AA 1
= I = A 1 A.
Formula for finding the inverse:
adjA1
Let A be a square matrix. Then A . =
det(A)
Theorem 1.6. If A and B are invertible matrices with the
1
same size, then AB is invertible and (AB) = B 1A 1.
Proof. We can establish the invertibility and obtain
the stated formula at the same time by showing that
(AB)(B 1 A 1 ) = (B 1 A 1 )(AB) = I. But (AB)(B 1 A 1 ) =
 17

A(BB 1 )A 1
= AIA 1
= AA 1
= I and similarly,
(B 1 A 1 )(AB) = I.
Remark 2: A product of any number of invertible matrices is
invertible, and the inverse of the product is the product of the
inverses in the reverse order.
Some properties:
If A is invertible and n is a nonnegative integer, then:
A0 = I and An = AA . . . A[nf actors]
n
A = (A 1 )n = A 1 A 1
. . . A 1 [nf actors]
Ar As = Ar+s and (Ar )s = Ars
1
A is invertible and (A 1 ) 1
= A.
An is invertible and (An ) 1
=A n
= (A 1 )n .
1
kA is invertible for any nonzero scalar k, and (kA) = k 1A 1.
Properties of the transpose:
If the sizes of the matrices are such that the stated operations
can be performed, then:
(AT )T = A
(A + B)T = AT + B T
(A B)T = AT BT
(kA)T = kAT
18

(AB)T = B T AT
Remark 3: The transpose of a product of any number of
matrices is the product of the transposes in the reverse order.
Theorem 1.7. If A is an invertible matrix, then AT is also
invertible and (AT ) 1
= (A 1 )T .
Proof. We can establish the invertibility and obtain the formula
at the same time by showing that AT (A 1 )T = (A 1 )T AT = I
But from the facts that (AB)T = B T AT and I T = I, we have
AT (A 1 )T = (A 1 A)T = I T = I
(A 1 )T AT = (AA 1 )T = I T = I which completes the proof.

Elementary row operations:

1. Multiply a row by a nonzero constant c.
2. Interchange two rows.
3. Add a constant c times one row to another.
It should be evident that if we let B be the matrix that results
from A by performing one of the operations in this list, then
the matrix A can be recovered from B by performing the
corresponding operation in the following list:
1. Multiply the same row by 1/c.
 19

2. Interchange the same two rows.

3. If B resulted by adding c times row ri of A to row rj , then
add -c times rj to ri .
It follows that if B is obtained from A by performing a
sequence of elementary row operations, then there is a second
sequence of elementary row operations, which when applied to
B recovers A.
Matrices A and B are said to be row equivalent if either
(hence each) can be obtained from the other by a sequence of
elementary row operations.
A matrix E is called an elementary matrix if it can be
obtained from an identity matrix by performing a single
elementary row2operation.
3
61 0 7
Consider I2 = 4 5. Multiplying the second row of I2 by 3
0 1
2 3
61 0 7
we get an elementary matrix 4 5.
0 3
Interchaning second and fourth rows of I4 yields an elementary
20
2 3
61 0 0 07
6 7
60 0 0 17
6 7
matrix 6 7.
60 0 1 07
6 7
4 5
0 1 0 0
Theorem 1.8.Row Operations by Matrix Multiplication: If
the elementary matrix E results from performing a certain row
operation on Im and if A is an m ⇥ n matrix, then the product
EA is the matrix that results when this same row operation is
performed on A.
Theorem 1.9. Every elementary matrix is invertible, and the
inverse is also an elementary matrix.
Proof. If E is an elementary matrix, then E results by perform-
ing some row operation on I . Let E0 be the matrix that results
when the inverse of this operation is performed on I . Applying
the theorem 1.8. and using the fact that inverse row opera-
tions cancel the e↵ect of each other, it follows that E0 E = I
and EE0 = I. Thus, the elementary matrix E0 is the inverse of E.
 21

1.5 Diagonal and triangular matrices

A square matrix in which all the entries o↵ the main diagonal

are zero is called a diagonal
2 matrix.3

6 2 0 0 07 2 3
2 3 2 3
6 7 6 1 0 07
6 7
6 1 0 7 6 0 8 0 07 6 7 60 07
Example. 4 ,
5 6 7 6 0 1 07
, 6
7, 4 5
0 3 6 7
1 07 4 5
60 0 0 0
4 5 0 0 1
0 0 0 4
2 general n ⇥ n 3diagonal matrix D can be written as
A

6d1 0 . . . 0 7
6 7
60 d ... 0 7
6 2 7
6. . .. 7 (1).
6 .. .. .77
6
4 5
0 0 . . . dn
A diagonal matrix is invertible if and only if all of its diagonal
entries
2 are nonzero; in this
3 case the inverse of (1) is
61/d1 0 ... 0 7
6 7
6 0 1/d2 . . . 0 7
6 7 1
6 . .. 7
.. 7 = D (2).
6 .. . . 7
6
4 5
0 0 . . . 1/dn
If D is the diagonal matrix (1) and k is a positive integer, then
22
2 3
k
6 1 0 ... 0 7
d
6 7
6 0 dk . . . 0 7
6 2 7 k
6. . 7
.. 7 = D (3).
6 .. .. .7
6
4 5
0 0 . . . dkn
2 3
61 0 07
6 7
Example. If A = 6 60 3 07 7, then
4 5
0 0 2
2 3 2 3
61 0 0 7 61 0 07
6 7 5 6 7
A 1 = 6 60 1/3 0 7 7 , A = 60
6 243 0 77,A
5
=
4 5 4 5
0 0 1/2 0 0 32
2 3
61 0 0 7
6 7
60 1/243 0 7
6 7
4 5
0 0 1/32

A square matrix in which all the entries above the main

diagonal are zero is called lower triangular, and a square
matrix in which all the entries below the main diagonal are
zero is called upper triangular. A matrix that is either upper
triangular or lower triangular is called triangular.
Observe that diagonal matrices are both upper triangular and
 23

lower triangular since they have zeros below and above the main

diagonal. Observe also that a square matrix in row echelon form is

upper triangular since it has zeros below the main diagonal.

Theorem 1.10.(a) The transpose of a lower triangular matrix

is upper triangular, and the transpose of an upper triangular
matrix is lower triangular.
(b) The product of lower triangular matrices is lower trian-
gular, and the product of upper triangular matrices is upper
triangular.
(c) A triangular matrix is invertible if and only if its diagonal
entries are all nonzero.
(d) The inverse of an invertible lower triangular matrix is lower
triangular, and the inverse of an invertible upper triangular
matrix is upper triangular.
Example.
2 Consider
3 the 2
upper triangular
3 matrices
61 3 17 63 2 27
6 7 6 7
6 7
A = 60 2 4 7 , B = 60 0 6 17
7
4 5 4 5
0 0 5 0 0 1
It follows from part (c) of the above theorem that the matrix
A is invertible but the matrix B is not. Moreover, the theorem
also tells us that A 1 , AB, and BA must be upper triangular.
24
2 3 2 3
61 3/2 7/5 7 63 2 27
6 7 6 7
Also, A 1 = 6
60 1/2 7
2/57 , AB = 60
6 0 277 , BA =
4 5 4 5
0 0 1/5 0 0 5
2 3
63 5 17
6 7
60 0 57
6 7
4 5
0 0 5

Remark. If A is an n ⇥ n triangular matrix (upper

triangular, lower triangular, or diagonal), then det(A) is the
product of the entries on the main diagonal of the matrix; that
is, det(A) = a11 a22 . . . ann .

A square2matrix A 3is said to be symmetric if A = AT .

1 4 5 2 3
6 7
6 7 67 37
Example. 64 6 7
3 07 , 4 5
4 5 3 5
5 0 7
All diagonal matrices are symmetric.

Algebraic properties of symmetric matrices

If A and B are symmetric matrices with the same size, and if
k is any scalar, then:
 25

(a) AT is symmetric.
(b) A + B and A - B are symmetric.
(c) kA is symmetric.
Theorem 1.11. The product of two symmetric matrices is
symmetric if and only if the matrices commute.
Proof. Let A and B be symmetric matrices with the same
size. Then it follows from the result (AB)T = BT AT and the
symmetry of A and B that
(AB)T = BT AT = BA.
Thus, (AB)T = AB if and only if AB = BA, that is, if and
only if A and B commute.
In general, a symmetric matrix need not be invertible. For
example, a diagonal matrix with a zero on the main diagonal is
symmetric but not invertible. But if a symmetric matrix hap-
pens to be invertible, then its inverse must also be symmetric.
Theorem 1.12. If A is an invertible symmetric matrix, then
1
A is symmetric.
Proof Assume that A is symmetric and invertible. From the
facts that
(AT ) 1
= (A 1 )T and A = AT , we have
26

(A 1 )T = (AT ) 1
=A 1

1
which proves that A is symmetric.
Remark. Matrix products of the form AAT and AT A arise in a
variety of applications. If A is an m ⇥ n matrix, then AT is an
n ⇥ m matrix, so the products AAT and AT A are both square
matrices—the matrix AAT has size m ⇥ m, and the matrix
AT A has size n ⇥ n. Such products are always symmetric since
(AAT )T = (AT )T AT = AAT and (AT A)T = AT (AT )T = AT A.
Example. One can verify that AAT and AT A are symmetric
for the
2 matrix 3
61 2 47
A=4 5
3 0 5
Theorem 1.13. If A is an invertible matrix, then AAT and
AT A are also invertible.
Proof Since A is invertible, so is AT by Theorem 1.7. Thus
AAT and AT A are invertible, since they are the products of
invertible matrices.
 CHAPTER 2

SYSTEM OF LINEAR EQUATIONS

2.1 Introduction

A system of linear equations is a fundamental concept in engineering mathe-

matics that involves multiple linear equations with common variables. These

systems appear in various scientific and engineering applications, such as electri-

cal circuit analysis, structural mechanics, and data modeling. By understanding

and solving these systems, engineers can predict behaviors, optimize solutions,

and analyze complex interactions in physical and computational models. The

study of linear equations forms the basis for more advanced mathematical tech-

niques, including matrix algebra and numerical methods, making it a crucial

topic engineering students.

Solving systems of linear equations allows to determine unknown variables

and establish relationships between different parameters. These equations of-

ten arise in practical scenarios such as balancing forces in mechanical systems,

analyzing fluid dynamics, and optimizing resource allocation in industrial pro-

cesses. The ability to systematically approach and solve these problems en-

hances problem-solving skills and mathematical reasoning, making it an essen-

tial area of study in engineering disciplines. Various methods exist to solve these

systems, ranging from simple substitution and elimination techniques to more

advanced matrix-based approaches like Gaussian elimination and Cramer’s rule.

In modern engineering, computational tools such as MATLAB and Python are

extensively used to handle large-scale linear systems efficiently. Mastery of these

techniques not only equips students with essential mathematical skills but also

prepares them for real-world problem-solving in diverse engineering fields.

 28

2.2 Introduction to system of linear

equations

In two dimensions a line in a rectangular xy-coordinate

system can be represented by an equation of the form
ax + by = c(a, b not both 0) and in three dimensions a plane
in a rectangular xyz-coordinate system can be represented by
an equation of the form ax + by + cz = d(a, b, c not all 0).
These are examples of “linear equations,” the first being a
linear equation in the variables x and y and the second a linear
equation in the variables x, y, and z.
More generally, we define a linear equation in the n variables
x1 , x2 , . . . , xn to be one that can be expressed in the form
a1 x1 + a2 x2 + . . . + an xn = b (1) where a1 , a2 , . . . , an and b are
constants, and the a0 s are not all zero. In the special cases where
n = 2 or n = 3, we will often use variables without subscripts
and write linear equations as a1 x + a2 y = b(a1 , a2 not both 0)
a1 x + a2 y + a3 z = b(a1 , a2 , a3 not all 0). In the
special case where b = 0, Equation (1) has the form
a1 x1 + a2 x2 + . . . + an xn = 0 which is called a homogeneous
29

linear equation in the variables x1 , x2 , . . . , xn .

A finite set of linear equations is called a system of linear
equations or, more briefly, a linear system. The variables
are called unknowns. A general linear system of m equations
in the n unknowns x1 , x2 , . . . , xn can be written as
a11 x1 + a12 x2 + . . . + a1n xn = b1
a21 x1 + a22 x2 + . . . + a2n xn = b2
... ... ... ... (2)
am1 x1 + am2 x2 + . . . + amn xn = bm .
The above system of equations can be expressed as a single
matrix
 equation AX = B (3), where
   
 a11 a12 . . . a1n   x1   b1 
     
a
 21 a22 . . . a2n 
 x  b 
 2  2
A= , X =  . , B =  . 
... ... ... ...  ..   .. 
     
     
am1 am2 . . . amn xn bn
Any column vector X which satisfies the matrix equation (3)
is called the solution of the system. That is X is the column
vector consisting of the set of values of x1 , x2 , . . . , xn which
satisfy simultaneously the m equations in the system (2).
When B 6= 0, the system is said to be non-homogeneous.
When B = 0, the system is said to be homogeneous.
 30

A linear system may have exactly one solution, an infinite

number of solutions or no solution at all. Systems that have
one or more solutions are called consistent system and
systems that do not have a solution are called inconsistent
systems.

2.3 Linear system in two and three un- knowns

Linear systems in two unknowns arise in connection with in-

tersections of lines. For example, consider the linear system

a1 x + b1 y = c1 , a2 x + b2 y = c2 in which the graphs of the

equations are lines in the xy-plane. Each solution (x, y) of this

system corresponds to a point of intersection of the lines, so there

are three possibilities:

1. The lines may be parallel and distinct, in which case there is no

intersection and consequently no solution.

2. The lines may intersect at only one point, in which case the

system has exactly one solution.

3. The lines may coincide, in which case there are infinitely

31

many points of intersection (the points on the common line) and

consequently infinitely many solutions.

Thus, a consistent linear system of two equations in two

unknowns has either one solution or infinitely many solu-

tions—there are no other possibilities. The same is true

for a linear system of three equations in three unknowns

a1 x + b1 y + c1 z = d1 , a2 x + b2 y + c2 z = d2 , a3 x + b3 y + c3 z = d3

in which the graphs of the equations are planes. The solutions

of the system, if any, correspond to points where all three planes

intersect, so again we see that there are only three possibilities—no

solutions, one solution, or infinitely many solutions.

Every system of linear equations has zero, one,

or infinitely many solutions. There are no other
possibilities.
Example 1. Solve the linear system
x−y =1
2x + y = 6.
Solution. We can eliminate x from the second equation by
using the first equation. This gives the system
x−y =1
 32

3y = 4.
From the second equation we obtain y = 4/3 and on substitut-
ing this value in the first equation we obtain x = 1 + y = 7/3.
Thus, the system has the unique solution x = 7/3, y = 4/3.
Geometrically, this means that the lines represented by the equa-

tions in the system intersect at the single point (7/3, 4/3).

Example 2. Solve the linear system

x+y =4
3x + 3y = 6.
Solution. We can eliminate x from the second equation by
using the first equation. This yields the simplified system
x+y =4
0 = −6.
The second equation is contradictory, so the given system has
no solution.
Geometrically, this means that the lines corresponding to the

equations in the original system are parallel and distinct.

Example 3. Solve the linear system

4x − 2y = 1
16x − 8y = 4.
33

Solution. We can eliminate x from the second equation by

using the first equation. This yields the system
4x − 2y = 1
0 = 0.
The second equation does not impose any restrictions on x and
y and hence can be omitted. Thus, the solutions of the system
are those values of x and y that satisfy the single equation
4x − 2y = 1.
Geometrically, this means the lines corresponding to the two

equations in the original system coincide.

One way to describe the solution set is to solve this equation

1
for x in terms of y to obtain x = (1 + 2y)/4, i.e; x = 4
+ 12 y
and then assign an arbitrary value t (called a parameter) to y.
1
Thus, we obtain x = 4
+ 12 t, y = t. This gives infinite number
of solutions for the given system.
Example 4. Solve the linear system
x − y + 2z = 5
2x − 2y + 4z = 10
3x − 3y + 6z = 15
Solution. This system can be solved by inspection, since
 34

the second and third equations are multiples of the first.

Geometrically, this means that the three planes coincide and that those
values of x, y, and z that satisfy the equation x − y + 2z = 5
automatically satisfy all three equations. Thus, it is enough to
find the solutions of x − y + 2z = 5.
We can do this by first solving this equation for x in terms of
y and z, then assigning arbitrary values r and s (parameters) to
y and z respectively, and then expressing the solution by the
three parametric equations
x = 5 + r − 2s, y = r, z = s.
Specific solutions can be obtained by choosing numerical values
for the parameters r and s. For example, taking r = 1 and
s = 0 yields the solution (6, 1, 0).

2.4 Non-homogeneous system

Consider the following m equations in n unknowns.

a11 x1 + a12 x2 + . . . + a1n xn = b1
a21 x1 + a22 x2 + . . . + a2n xn = b2
... ... ... ...
35

am1 x1 + am2 x2 + . . . + amn xn = bm .

The above system of equations can be expressed as a single
matrix
 equation AX = B, where    
 a11 a12 . . . a1n   x1   b1 
     
a a . . . a  x  b 
21 22 2n 2  2
A= , X =  . , B =  . 
   
... ... ... ...  ..   .. 
     
     
am1 am2 . . . amn xn bm
The matrix [AB], obtained by placing the column matrix B to
the right of the matrix A is called the augmented matrix for
the system.
 
 a11 a12 . . . a1n b1 
 
a
 21 a22 . . . a2n b2 
AB =  .

... ... ... ... . . .
 
 
am1 am2 . . . amn bm
The algebraic operations to perform to solve the system are:
1. Multiply an equation through by a nonzero constant.
2. Interchange two equations.
3. Add a constant times one equation to another.
Correspondingly these are the following operations to perform
to solve the system on the rows of the augmented matrix:
1. Multiply a row through by a nonzero constant.
 36

2. Interchange two rows.

3. Add a constant times one row to another.
These are called elementary row operations on a matrix.

Row reduced echelon form

By a sequence of elementary row operations, the augumented
matrix of a linear system can be transformed to a matrix that
have the following properties:
1. If a row does not consist entirely of zeros, then the first
nonzero number in the row is a 1. We call this a leading 1.
2. If there are any rows that consist entirely of zeros, then
they are grouped together at the bottom of the matrix.
3. In any two successive rows that do not consist entirely of
zeros, the leading 1 in the lower row occurs farther to the right
than the leading 1 in the higher row.
4. Each column that contains a leading 1 has zeros everywhere
else in that column.
A matrix that has the first three properties is said to be in
row reduced echelon form.
Algorithm for reducing to row echelon form
37

Step 1. Locate the leftmost column that does not consist entirely of zeros.

Step 2. Interchange the top row with another row, if necessary, to bring a

nonzero entry to the top of the column found in Step 1.

Step 3. If the entry that is now at the top of the column found in Step 1

is a, multiply the first row by 1/a in order to introduce a leading 1.

Step 4. Add suitable multiples of the top row to the rows below so that

all entries below the leading 1 become zeros.

Step 5. Now cover the top row in the matrix and begin again with Step

1 applied to the submatrix that remains. Continue in this way until the

entire matrix is in row echelon form.

Step 6. Beginning with the last nonzero row and working upward, add

suitable multiples of each row to the rows above to introduce zeros above

the leading 1’s.  

1 2 1 
 
3 1 −2
Example 5. Reduce the matrix A =   to echelon
 
4 −3 −1
 
 
2 4 2
form.
 38
 
1 2 1
 
3 1 −2
Solution. A = 
 

4 −3 −1
 
 
2 4 2
R2→ R2 − 3R1 , R3 → R3 − 4R1 , R4 → R4 − 2R1

1 2 1
 
0 −5 −5
∼
=
 

0 −11 −5
 
 
0 0 0
R2 → −1 R
 5 2 
1 2 1
 
0 1 1 
∼
=
 

0 −11 −5
 
 
0 0 0
R3→ R3 + 11R2

1 2 1
 
0 1 1
∼
=
 

0 0 6
 
 
0 0 0
R3 → 16 R3
39
 
1 2 1
 
0 1 1
∼
=

.

0 0 1
 
 
0 0 0

2.5 Gauss Elimination method for

solving a system of linear equa-
tions

Example 6. Solve the below system by Gauss elimination

method.
x + 2y + z = 2
3x + y − 2z = 1
4x − 3y − z = 3
2x + 4y + 2z = 4.    
1 2 1 2
   
3 1 −2 1
Solution. Here A =  , B =  .
   

4 −3 −1 3
   
   
2 4 2 4
 40
 
1 2 1 2
 
3 1 −2 1
So augmented matrix [AB] =  .
 
4 −3 −1 3
 
 
2 4 2 4
Reduce the [AB] torow echelonform.

1 2 1 2
 
0 1 1 1
That gives [AB] =  . (do as in example 5.)
 
0 0 1 1
 
 
0 0 0 0
The
 corresponding
 system
 of equations in matrix form is
1 2 1   2
  x  
3 1 −2  1
 
 y  =   , which is equivalent to the equa-
    

4 −3 −1   1
   
  z  
2 4 2 0
tions x + 2y + z = 2, y + z = 1, z = 1.
By back substitution, we get the solution as x = 1, y = 0, z = 1.
Example 7. Solve the below system by Gauss elimination
method.
−2x3 + 7x5 = 12
2x1 + 4x2 − 10x3 + 6x4 + 12x5 = 28
2x1 + 4x2 − 5x3 + 6x4 − 5x5 = −1.
41

Solution.
 Here the augmented matrix
  is 
0 0 −2 0 7 12  2 4 −10 6 12 28 
∼ ∼
   
[AB] = 2 4 −10 6 12 28  = 0 0 −2 0 7 12  =
   
2 4 −5 6 −5 −1 2 4 −5 6 −5 −1
   
1 2 −5 3 6 14  1 2 −5 3 6 14 
0 0 −2 0 7 12  ∼ ∼
   
 = 0 0 −2 0 7  =
12 


   
2 4 −5 6 −5 −1 0 0 5 0 −17 −29
   
1 2 −5 3 6 14  1 2 −5 3 6 14 
0 0 1 0 −7/2 −6  ∼ 0 0 1 0 −7/2 −6  ∼
   
 =
  =
   
0 0 5 0 −17 −29 0 0 5 0 −17 −29
   
1 2 −5 3 6 14  1 2 −5 3 6 14 
0 0 1 0 −7/2 −6 ∼
   
= 0 0 1 0 −7/2 −6,
   
   
0 0 0 0 1/2 1 0 0 0 0 1 2
which is now in echelon form.
Corresponding matrix form is
 
x
   1  
 
1 2 −5 3 6  x2   14 

    
0 0 1 0 −7/2 x  = −6 which is equivalent to
   3  
    
0 0 0 0 1 x4  2
 
 
x5
x1 + 2x2 − 5x3 + 3x4 + 6x5 = 14
 42

x3 − 72 x5 = −6
x5 = 2. By back substitution, x5 = 2, x3 = 1.
Assigning x2 = r, x4 = s, the solution is
x1 = 7 − 2r − 3s, x2 = r, x3 = 1, x4 = s, x5 = 2.
2.6 Gauss Jordan elimination method for
solving a system of linear equations .
Example 8. Solve the below system by Gauss Jordan
elimination method.
x1 + 3x2 − 2x3 + 2x5 = 0
2x1 + 6x2 − 5x3 − 2x4 + 4x5 − 3x6 = −1
5x3 + 10x4 + 15x6 = 5
2x1 + 6x2 + 8x4 + 4x5 + 18x6 = 6
Solution.
 Here the augmented matrix
 is 
1 3 −2 0 2 0 0  1 3 −2 0 2 0 0
   
2 6 −5 −2 4 −3 −1 0 0 −1 −2 0 −3 −1
∼ ∼
[AB] =  = =

0 0 5 10 0 15 5  0 0 5 10 0 15 5 
   
   
2 6 0 8 4 18 6 0 0 4 8 0 18 6
   
1 3 −2 0 2 0 0 1 3 −2 0 2 0 0
   
0 0 1 2 0 3 1
 ∼ 0 0 1 2 0 3 1

 ∼
 =   =


0 0 5 10 0 15 5 0 0 0 0 0 0 0
   
   
0 0 4 8 0 18 6 0 0 0 0 0 6 2
43
   
1 3 −2 0 2 0 0 1 3 −2 0 2 0 0 
   
0 0 1 2 0 3 1 0 0 1 2 0 3 1 
 ∼  ∼
 =  =
 
 
0 0 0 0 0 6 2 0 0 0 0 0 1 1/3
   
   
0 0 0 0 0 0 0 0 0 0 0 0 0 0
   
1 3 −2 0 2 0 0  1 3 0 4 2 0 0 
   
0 0 1 2 0 0 0  0 0 1 2 0 0 0 
∼
= , which
  
 
0 0 0 0 0 1 1/3 0 0 0 0 0 1 1/3
   
   
0 0 0 0 0 0 0 0 0 0 0 0 0 0
is now in echelon form.
Corresponding matrix form
 is
x1 
    
1 3 0 4 2 0 
x  0 
2
   
0 0 1 2 0 0    

x
  3
  0 

=  which is equivalent to
 
 
0 0 0 0 0 1  x 1/3
  
4  
    
0 0 0 0 0 0  x5 
 0
 
x6
x1 + 3x2 + 4x4 + 2x5 = 0
x3 + 2x4 = 0
x6 = 1/3.
Solving , we obtain x1 = −3x2 − 4x4 − 2x5 x3 = −2x4 x6 = 1/3
Finally, by assigning the free variables x2 , x4 , x5 arbitrary
 44

values r, s, t, respectively, we get

x1 = −3r − 4s − 2t, x2 = r, x3 = −2s, x4 = s, x5 = t, x6 = 1.
Example 9. Solve the below system by Gauss Jordan
elimination method.
x1 + 2x2 − 3x3 − 4x4 = 6
x1 + 3x2 + x3 − 2x4 = 4
2x1 + 5x2 − 2x3 − 5x4 = 10
Solution.
 Here the augmented
 matrix is
1 2 −3 −4 6 
 
[AB] = 1 3 1 −2 4 . The row reduced echelon form

 
2 5 −2 −5 10
is  
1 0 −11 0 10 
[AB] ∼
 
= 0 1 4 0 −2.
 
 
0 0 0 1 0
Corresponding matrix
  form is
  x  
 1
1 0 −11 0    10 
  x   
2
0 1 4 0  
  =−2 which is equivalent to

x3 

   
0 0 0 1   0
x4
x1 − 11x3 = 10
x2 + 4x3 = −2
45

x4 = 0.
Assinging x3 = t, we get the general solution as
x1 = 11t + 10, x2 = −4t − 2, x3 = t, x4 = 0.
Example 10. Suppose that the matrices below are augmented
matrices for linear systems in the unknowns x1 , x2 , x3 , x4 .
Discuss the existence and uniqueness of solutions to the
corresponding
 linear systems.  
1 −3 7 2 5 1 −3 7 2 5
   
0 1 2 −4 1 0 1 2 −4 1
A =   B =
   
 
0 0 1 6 9 0 0 1 6 9
   
   
0 0 0 0 1 0 0 0 0 0
 
1 −3 7 2 5
 
0 1 2 −4 1
C= .
 
0 0 1 6 9
 
 
0 0 0 1 0
 46

2.7 Homogeneous system of linear

equations

A system of linear equations is said to be homogeneous if the

constant terms are all zero; that is, the system has the form
a11 x1 + a12 x2 + . . . + a1n xn = 0
a21 x1 + a22 x2 + . . . + a2n xn = 0
... ... ... ... (2)
am1 x1 + am2 x2 + . . . + amn xn = 0.
Here B = 0, so the matrix A and the augmented matrix
[AB] are the same. Every homogeneous system of lin-
ear equations is consistent because all such systems have
x1 = 0, x2 = 0, ..., xn = 0 as a solution. This solution is called
the trivial solution; if there are other solutions, they are called
nontrivial solutions.
Because a homogeneous linear system always has the trivial
solution, there are only two possibilities for its solutions:
ˆ The system has only the trivial solution.

ˆ The system has infinitely many solutions in addition to the

trivial solution.
47

Theorem 2.1 Free Variable Theorem for Homogeneous

Systems If a homogeneous linear system has n unknowns, and

if the reduced row echelon form of its augmented matrix has r
nonzero rows, then the system has n - r free variables.
Theorem 2.2 A homogeneous linear system with more
unknowns
than equations has infinitely many solutions.
Proof. If a homogeneous linear system has m equations in n
unknowns, and if m < n, then it must also be true that r <
n. This being the case, the theorem implies that there is at
least one free variable, and this implies that the system has
infinitely many solutions.
Example 11. Solve the below system .
x1 + 3x2 − 2x3 + 2x5 = 0
2x1 + 6x2 − 5x3 − 2x4 + 4x5 − 3x6 = 0
5x3 + 10x4 + 15x6 = 0
2x1 + 6x2 + 8x4 + 4x5 + 18x6 = 0
Solution. Here the augmented matrix is
 48
 
1 3 −2 0 2 0 0
 
2 6 −5 −2 4 −3 0
[AB] =  . Continuing as in exam-
 
0 0 5 10 0 15 0
 
 
2 6 0 8 4 18 0
ple
 8., we get the echelon
 form of this matrix as

1 3 0 4 2 0 0
 
0 0 1 2 0 0 0
. Corresponding system of equations is
 

0 0 0 0 0 1 0
 
 
0 0 0 0 0 0 0
x1 + 3x2 + 4x4 + 2x5 = 0
x3 + 2x4 = 0
x6 = 0.
Solving , we obtain x1 = −3x2 − 4x4 − 2x5 x3 = −2x4 x6 = 0.
Finally, by assigning the free variables x2 , x4 , x5 arbitrary
values r, s, t, respectively, we get x1 = −3r − 4s − 2t, x2 =
r, x3 = −2s, x4 = s, x5 = t, x6 = 0.
Example 12. Solve the below system .
x1 − 2x2 + x3 − x4 = 0
x1 + x2 − 2x3 + 3x4 = 0
4x1 + x2 − 5x3 + 8x4 = 0
5x1 − 7x2 + 2x3 − x4 = 0
49

Solution. After reduucing to echelon form the corresponding

system
 of equations
  inthe matrix
 form will be

1 0 −1 5/3 x1  0

    
0 1 −1 4/3 x  0
  2  
  =  


0 0 0 0 
 x3  0
   

    
0 0 0 0 x4 0
Solving this and by substituting x3 = r, x4 = s, we get infinite
number of solutions.
The general solution is x1 = r − 53 s, x2 = r − 43 s, x3 = r, x4 = s.

2.8 More on linear systems

Theorem 2.3. A system of linear equations has zero, one,

or infinitely many solutions. There are no other possibilities.

Proof. If Ax = b is a system of linear equations, exactly one
of the following is true:
(a) the system has no solutions, (b) the system has exactly
one solution, or (c) the system has more than one solution.
The proof will be complete if we can show that the system has
infinitely many solutions in case (c).
 50

Assume that Ax = b has more than one solution, and let x0

= x1 - x2 , where x1 and x2 are any two distinct solutions.
Because x1 and x2 are distinct, the matrix x0 is nonzero.
Moreover, Ax0 = A(x1 - x2 ) = Ax1 - Ax2 = b - b = 0.
If we now let k be any scalar, then
A(x1 + kx0 ) = Ax1 + A(kx0 ) = Ax1 + k(Ax0 ) = b + k0 = b
+ 0 = b.
But this says that x1 + kx0 is a solution of Ax = b. Since x0 is
nonzero and there are infinitely many choices for k, the system
Ax = b has infinitely many solutions.
Solving Linear Systems by Matrix Inversion
Theorem 1.16. If A is an invertible n × n matrix, then for
each n × 1 matrix b, the system of equations Ax = b has
exactly one solution, namely, x = A−1 b.
Proof. Since A(A−1 b) = b, it follows that x = A−1 b is a
solution of Ax = b.
To show that this is the only solution, we will assume that x0
is an arbitrary solution and then show that x0 must be the
solution A−1 b.
If x0 is any solution of Ax = b, then Ax0 = b. Multiplying
51

both sides of this equation by A−1 , we obtain x0 = A−1 b.

Theorem 1.17. Equivalent Statements
If A is an n × n matrix, then the following statements are
equivalent:
(a) A is invertible.
(b) Ax = 0 has only the trivial solution.
(c) The reduced row echelon form of A is In .
(d) A is expressible as a product of elementary matrices.
Proof.
(a) ⇒ (b) Assume A is invertible and let x0 be any solution of
Ax = 0. Multiplying both sides of this equation by the matrix
A−1 gives A−1 (Ax0 ) = A−1 0, or (A−1 A)x0 = 0, or Ix0 = 0, or
x0 = 0. Thus, Ax = 0 has only the trivial solution.
(b) ⇒ (c) Let Ax = 0 be the matrix form of the system
a11 x1 + a12 x2 + . . . + a1n xn = 0
a21 x1 + a22 x2 + . . . + a2n xn = 0
... ... ... ... (1)
an1 x1 + an2 x2 + . . . + ann xn = 0.
and assume that the system has only the trivial solution. If
we solve by Gauss–Jordan elimination, then the system of
 52

equations corresponding to the reduced row echelon form of

the augmented matrix will be
x1 =0
x2 =0
..
. (2)
xn = 0.  
 a11 a12 . . . a1n 0 
 
a a . . . a 0 
21 22 2n
Thus the augmented matrix 
 

 . . . . . . . . . . . . . . .
 
 
an1 an2 . . . ann 0
for
 (1) can be reduced 
to the augmented matrix

 1 a12 . . . 0 0 
 
0 1 . . . 0 0 
 for (2) by a sequence of elementary
 

. . . . . . . . . . . . . . .
 
 
0 0 ... 1 0
row operations. If we disregard the last column (all zeros) in
each of these matrices, we can conclude that the reduced row
echelon form of A is In .
(c) ⇒ (d) Assume that the reduced row echelon form of A
is In , so that A can be reduced to In by a finite sequence of
elementary row operations. By theorem 1.8, each of these
53

operations can be accomplished by multiplying on the left

by an appropriate elementary matrix. Thus we can find
elementary matrices E1 , E2 ,...,Ek such that
Ek ··· E2 E1 A = In (3).
By theorem 1.9, E1 , E2 ,...,Ek are invertible. Multiplying both
sides of Equation (3) on the left successively by E−1
k
,...,E−1
2
,
E−1
1
we obtain
A = E−1
1
E−1
2
··· E−1
k
In = E−1
1
E−1
2
··· E−1
k
(4).
By theorem 1.9, this equation expresses A as a product of
elementary matrices.
(d)⇒ (a) If A is a product of elementary matrices, then from
the remark 2, properties and theorem 1.9, the matrix A is a
product of invertible matrices and hence is invertible.
A Method for Inverting Matrices
An application of Theorem 1.17.
Inversion Algorithm:
To find the inverse of an invertible matrix A, find a sequence
of elementary row operations that reduces A to the identity
and then perform that same sequence of operations on In to
obtain A−1 .
 54
 
1 2 3
 
Example: Find the inverse of 2 5 3.

 
1 0 8
We want to reduce A to the identity matrix by row operations
and simultaneously apply these operations to I to produce
A−1 . To accomplish this we will adjoin the identity matrix to
the right side of A, thereby producing a partitioned matrix
of the form [A|I]. Then we will apply row operations to this
matrix until the left side is reduced to I ; these operations will
convert the right side to A−1 , so the final matrix will have the
form
 [I|A−1 ]. Now the
 computation:

 1 2 3 1 0 0 
 
 2 5 3 0 1 0 
 
 
1 0 8 0 0 1
 
 1 2 3 1 0 0 
 
 0 1 −3 −2 1 0 
 
 
0 −2 5 −1 0 1
We added -2 times the first row to the second and -1 times the
first row to the third.
55
 
 1 2 3 1 0 0 
 
 0 1 −3 −2 1 0 
 
 
0 0 −1 −5 2 1
We added 2 times the second
  row to the third.
 1 2 3 1 0 0 
 
 0 1 −3 −2 1 0 
 
 
0 0 1 5 −2 −1
We
 multiplied the third row
 by -1.
 1 2 0 −14 6 3 
 
 0 1 0 13 −5 −3 
 
 
0 0 1 5 −2 −1
We added 3 times the third row to the second and -3 times the
third row to the first.
 
 1 0 0 −40 16 9 
 
 0 1 0 13 −5 −3 
 
 
0 0 1 5 −2 −1
We added -2 times
 the second row to the first.
−40 16 9 
 
Thus, A−1 =   13 −5 −3 .

 
5 −2 −1
Example. Consider the system of linear equations
x1 + 2x2 + 3x3 = 5
 56

2x1 + 5x2 + 3x3 = 3

x1 + 8x3 = 17
In matrix
 form this system
 can be
 written
 asAx= b, where

 1 2 3 1 0 0  x1  5
     
A=  2 5 3 0 1 0  , X = x2  , B =  3 
    
     
1 0 8 0 0 1 x3 17
In the above
 example, we showed that A is invertible and

−40 16 9
 
A−1 =  13 −5 −3 .
 
5 −2 −1
Thus, the solution
 of the system
  is  
−40 16 9   5   1 
    
x = A−1 b = 
 13 −5 −3  3  = −1.
   
    
5 −2 −1 17 2
Hence, x1 = 1, x2 = −1, x3 = 2.
Example:Show that the given below matrix is not invertible.

1 6 4 
 
2 4 −1.
 
 
1 2 5
Proceeding
57
 
 1 6 4 1 0 0 
 
 2 4 −1 0 1 0 
 
 
1 2 5 0 0 1
 
 1 6 4 1 0 0 
 
 0 −8 −9 −2 1 0 
 
 
0 8 9 1 0 1
We added -2 times the first row to the second and added the
first row to the third.
 
 1 6 4 1 0 0 
 
 0 −8 −9 −2 1 0 
 
 
0 0 0 −1 1 1
We added the second row to the third.
Since we have obtained a row of zeros on the left side, A is not
invertible.
Example: Use Theorem 1.17. to determine whether the given
homogeneous systems have nontrivial solutions.
(a) x1 + 2x2 + 3x3 = 0
2x1 + 5x2 + 3x3 = 0
x1 + 8x3 = 0
(b) x1 + 6x2 + 4x3 = 0
2x1 + 4x2 − x3 = 0
 58

− x1 + 2x2 + 5x3 = 0
Solution. From parts (a) and (b) of Theorem 1.17. a ho-
mogeneous linear system has only the trivial solution if and
only if its coefficient matrix is invertible. From the above
examples the coefficient matrix of system (a) is invertible and
that of system (b) is not. Thus, system (a) has only the trivial
solution while system (b) has nontrivial solutions.
Linear Systems with a Common Coefficient Matrix
One is concerned with solving a sequence of systems
Ax = b1 , Ax = b2 , Ax = b3 , ..., Ax = bk
each of which has the same square coefficient matrix A. If A
is invertible, then the solutions x1 = A−1 b1 , x2 = A−1 b2 , x3 =
A−1 b3 , ..., xk = A−1 bk
can be obtained with one matrix inversion and k matrix
multiplications. An efficient way to do this is to form the
partitioned matrix
 
A | b1 | b2 | . . . bk (1)
in which the coefficient matrix A is “augmented” by all k of
the matrices b1 , b2 , ..., bk , and then reduce (1) to reduced row
echelon form by Gauss–Jordan elimination. In this way we
59

can solve all k systems at once. This method has the added
advantage that it applies even when A is not invertible.
Example. Solve the systems
(a) x1 + 2x2 + 3x3 = 4 2x1 + 5x2 + 3x3 = 5 x1 + 8x3 = 9
(b) x1 + 2x2 + 3x3 = 1 2x1 + 5x2 + 3x3 = 6 x1 + 8x3 = −6
Solution. The two systems have the same coefficient matrix.
If we augment this coefficient matrix with the columns of
constants on the right
  sides of these systems, we obtain
 1 2 3 4 1 
 
 2 5 3 5 6 
 
 
1 0 8 9 −6
Reducing
 this matrix
 to reduced row echelon form we get
 1 0 1 1 2 
 
 0 1 0 0 1 
 
 
0 0 1 1 −1
It follows from the last two columns that the solution of system
(a) is x1 = 1, x2 = 0, x3 = 1 and the solution of system (b) is
x1 = 2, x2 = 1, x3 = −1.
Theorem 1.18. Let A be a square matrix.
(a) If B is a square matrix satisfying BA = I, then B = A−1 .
(b) If B is a square matrix satisfying AB = I, then B = A−1 .
 60

Proof. (a) Assume that BA = I . If we can show that A is

invertible, the proof can be completed by multiplying BA = I
on both sides by A−1 to obtain
BAA−1 = IA−1 or BI = IA−1 or B = A−1 .
To show that A is invertible, it suffices to show that the system
Ax = 0 has only the trivial solution. Let x0 be any solution of
this system. If we multiply both sides of Ax0 = 0 on the left
by B, we obtain
BAx0 = B0 or Ix0 = 0 or x0 = 0.
Thus, the system of equations Ax = 0 has only the trivial
solution.
Similarly (b).
Theorem 1.19. Equivalent Statements
If A is an n × n matrix, then the following statements are
equivalent:
(a) A is invertible.
(b) Ax = 0 has only the trivial solution.
(c) The reduced row echelon form of A is In .
(d) A is expressible as a product of elementary matrices.
(e) Ax = b is consistent for every n × 1 matrix b.
61

(f) Ax = b has exactly one solution for every n × 1 matrix b

Proof.
Since we proved in Theorem 1.17 that (a), (b), (c), and (d )
are equivalent, it will be sufficient to prove that
(a) ⇒ (f) ⇒ (e) ⇒ (a).
(a) ⇒ (f) Since A(A−1 b) = b, it follows that x = A−1 b is a
solution of Ax = b.
To show that this is the only solution, we will assume that x0
is an arbitrary solution and then show that x0 must be the
solution A−1 b.
If x0 is any solution of Ax = b, then Ax0 = b. Multiplying
both sides of this equation by A−1 , we obtain x0 = A−1 b.
(f) ⇒ (e) This is almost self-evident, for if Ax = b has exactly
one solution for every n × 1 matrix b, then Ax = b is consistent
for every n × 1 matrix b.
(e) ⇒ (a) If the system Ax = b is consistent for every n × 1
matrix b, then, in particular, this is so for the systems
 62
     
1 0 0
     
     
0 1 0
     
     
Ax = 0 Ax = 0 . . . Ax = 0
    
 ..   ..   .. 
     
. . .
     
0 0 1
Let x1 , x2 , ..., xn be solutions of the respective systems, and let
us form an n × n matrix C having these solutions as columns.
Thus C has the form
 
C = x1 | x2 | . . . | xn .
The successive columns of the product AC will be
Ax1 , Ax2 , ..., Axn .Thus,
 
1 0 ... 0
 
 
 0 1 ... 0

 
 
AC = Ax1 | Ax2 | . . . | Axn =  0 0 ... 0 =I
 .. .. .. 
 
. . .
 
0 0 ... 1
By part (b) of the above theorem, it follows that C = A−1 .
Thus, A is invertible.
Theorem 1.20. Let A and B be square matrices of the same
size. If AB is invertible, then A and B must also be invertible.
Proof. We will show first that B is invertible by showing that
63

the homogeneous system Bx = 0 has only the trivial solution.

If we assume that x0 is any solution of this system, then
(AB)x0 = A(Bx0 ) = A0 = 0
so x0 = 0 by parts (a) and (b) of the above theorem applied to
the invertible matrix AB. But the invertibility of B implies the
invertibility of B−1 , which in turn implies that
(AB)B−1 = A(BB−1 ) = AI = A
is invertible since the left side is a product of invertible
matrices. This completes the proof.
To answer the problem: Let A be a fixed m × n matrix. Find
all m × 1 matrices b such that the system of equations Ax =
b is consistent, we have
if A is an invertible matrix, Theorem 1.16 completely solves
this problem by asserting that for every m × 1 matrix b, the
linear system Ax = b has the unique solution x = A−1 b. If
A is not square, or if A is square but not invertible, then
Theorem 1.16 does not apply. In these cases b must usually
satisfy certain conditions in order for Ax = b to be consistent.
Example. What conditions must b1 , b2 , and b3 satisfy in order
for the system of equations
 64

x1 + x2 + 2x3 = b1
x 1 + x 3 = b2
2x1 + x2 + 3x3 = b3 to be consistent?
Solution.
 The augmented matrix is

1 1 2 b1 
 
1 0 1 b 
 2
 
2 1 3 b3
which can be reduced 
 to row echelon form as follows:
1 1 2 b1 
 
0 −1 −1 b − b 
 2 1 
 
0 −1 −1 b3 − 2b1
 
1 1 2 b1 
 
0 −1 −1 b − b 
 1 2 
 
0 −1 −1 b3 − 2b1
 
1 1 2 b1 
 
0 −1 −1 b − b 
 1 2 
 
0 0 0 b3 − b2 − b1
From the third row, it is clear that the system has a solution if
and only if b1 , b2 , and b3 satisfy the condition b3 − b2 − b1 = 0 or
b3 = b1 + b2 . To express this condition another way, Ax = b is
65
 
 b1 
 
consistent if and only if b is a matrix of the form b = 
 b2 

 
b1 + b2
where b1 , b2 are arbitrary.
Example. What conditions must b1 , b2 , and b3 satisfy in order
for the system of equations
x1 + 2x2 + 3x3 = b1
2x1 + 5x2 + 3x3 = b2
1x1 + 8x3 = b3 to be consistent?
Solution.
 The augmented
 matrix is

 1 2 3 b1 
 
 2 5 3 b .
 2 
 
1 0 8 b3
Reducing
 this to reduced row echelon
 we get

 1 0 0 −40b1 16b2 9b3 

 
 0 1
 0 13b1 −5b2 −3b3  
 
0 0 1 5b1 −2b2 −1b3
In this case there are no restrictions on b1 , b2 , and b3 , so the
system has the unique solution x1 = −40b1 + 16b2 + 9b3 , x2 =
13b1 −5b2 −3b3 , x3 = 5b1 −2b2 −b3 for all values of b1 , b2 , and b3 .
 66
2.9. LU decomposition
Suppose we have the system of equations
AX = B.
The motivation for an LU decomposition is based on the observation that systems of equations
involving triangular coefficient matrices are easier to deal with. Indeed, the whole point of Gaussian
elimination is to replace the coefficient matrix with one that is triangular. The LU decomposition is
another approach designed to exploit triangular systems.
We suppose that we can write
A = LU
where L is a lower triangular matrix and U is an upper triangular matrix. Our aim is to find L and
U and once we have done so we have found an LU decomposition of A.

Key Point 5
An LU decomposition of a matrix A is the product of a lower triangular matrix and an upper
triangular matrix that is equal to A.

It turns out that we need only consider lower triangular matrices L that have 1s down the diagonal.
Here is an example. Let
     
1 2 4 1 0 0 U11 U12 U13
A =  3 8 14  = LU where L =  L21 1 0  and U =  0 U22 U23 .
2 6 13 L31 L32 1 0 0 U33
Multiplying out LU and setting the answer equal to A gives
   
U11 U12 U13 1 2 4
 L21 U11 L21 U12 + U22 L21 U13 + U23  =  3 8 14  .
L31 U11 L31 U12 + L32 U22 L31 U13 + L32 U23 + U33 2 6 13
Now we use this to find the entries in L and U . Fortunately this is not nearly as hard as it might at
first seem. We begin by running along the top row to see that

U11 = 1 , U12 = 2 , U13 = 4 .

Now consider the second row

L21 U11 = 3 ∴ L21 × 1 = 3 ∴ L21 = 3 ,

L21 U12 + U22 = 8 ∴ 3 × 2 + U22 = 8 ∴ U22 = 2 ,

L21 U13 + U23 = 14 ∴ 3 × 4 + U23 = 14 ∴ U23 = 2 .

 67®
Notice how, at each step, the equation being considered has only one unknown in it, and other
quantities that we have already found. This pattern continues on the last row

L31 U11 = 2 ∴ L31 × 1 = 2 ∴ L31 = 2 ,

L31 U12 + L32 U22 = 6 ∴ 2 × 2 + L32 × 2 = 6 ∴ L32 = 1 ,

L31 U13 + L32 U23 + U33 = 13 ∴ (2 × 4) + (1 × 2) + U33 = 13 ∴ U33 = 3 .

We have shown that

    
1 2 4 1 0 0 1 2 4
A =  3 8 14  =  3 1 0   0 2 2 
2 6 13 2 1 1 0 0 3
and this is an LU decomposition of A.

Task  
3 1
Find an LU decomposition of .
−6 −4

Answer
Let
      
3 1 1 0 U11 U12 U11 U12
= LU = =
−6 −4 L21 1 0 U22 L21 U11 L21 U12 + U22
then, comparing the left and right hand sides row by row implies that U11 = 3, U12 = 1, L21 U11 = −6
which implies L21 = −2 and L21 U12 + U22 = −4 which implies that U22 = −2. Hence
    
3 1 1 0 3 1
=
−6 −4 −2 1 0 −2
 
3 1
is an LU decomposition of .
−6 −4
 68
 
Task 3 1 6
Find an LU decomposition of  −6 0 −16 .
0 8 −17

Your solution

Answer
Using material from the worked example in the notes we set
   
3 1 6 U11 U12 U13
 −6 0 −16  =  L21 U11 L21 U12 + U22 L21 U13 + U23 
0 8 −17 L31 U11 L31 U12 + L32 U22 L31 U13 + L32 U23 + U33
and comparing elements row by row we see that
U11 = 3, U12 = 1, U13 = 6,
L21 = −2, U22 = 2, U23 = −4
L31 = 0 L32 = 4 U33 = −1
and it follows that
    
3 1 6 1 0 0 3 1 6
 −6 0 −16  =  −2 1 0   0 2 −4 
0 8 −17 0 4 1 0 0 −1
is an LU decomposition of the given matrix.
 69
®

2.9.1. Using LU decomposition to solve systems of equations

Once a matrix A has been decomposed into lower and upper triangular parts it is possible to obtain
the solution to AX = B in a direct way. The procedure can be summarised as follows

• Given A, find L and U so that A = LU . Hence LU X = B.

• Let Y = U X so that LY = B. Solve this triangular system for Y .

• Finally solve the triangular system U X = Y for X.

The benefit of this approach is that we only ever need to solve triangular systems. The cost is that
we have to solve two of them.
[Here we solve only small systems; a large system is presented later.]

Example 6       
x1 1 2 4 x1 3
Find the solution of X =  x2  of the system  3 8 14   x2  =  13  .
x3 2 6 13 x3 4

Solution

• The first step is to calculate the LU decomposition of the coefficient matrix on the left-hand
side. In this case that job has already been done since this is the matrix we considered earlier.
We found that
   
1 0 0 1 2 4
L =  3 1 0 , U =  0 2 2 .
2 1 1 0 0 3
 
y1
• The next step is to solve LY = B for the vector Y =  y2 . That is we consider
y3
    
1 0 0 y1 3
LY =  3 1 0   y2  =  13  = B
2 1 1 y3 4

which can be solved by forward substitution. From the top equation we see that y1 = 3.
The middle equation states that 3y1 + y2 = 13 and hence y2 = 4. Finally the bottom line
says that 2y1 + y2 + y3 = 4 from which we see that y3 = −6.
 70

Solution (contd.)

• Now that we have found Y we finish the procedure by solving U X = Y for X. That is we
solve
    
1 2 4 x1 3
UX =  0 2 2   x2  =  4  = Y
0 0 3 x3 −6

by using back substitution. Starting with the bottom equation we see that 3x3 = −6 so
clearly x3 = −2. The middle equation implies that 2x2 + 2x3 = 4 and it follows that x2 = 4.
The top equation states that x1 + 2x2 + 4x3 = 3 and consequently x1 = 3.

Therefore we have found that the solution to the system of simultaneous equations
      
1 2 4 x1 3 3
 3 8 14   x2  =  13  is X= 4 .
2 6 13 x3 4 −2

Task
Use the LU decomposition you found earlier in the last Task (page 24) to solve
    
3 1 6 x1 0
 −6 0 −16   x2  =  4 .
0 8 −17 x3 17

Your solution
 71 ®

Answer   
1 0 0 3 1 6
We found earlier that the coefficient matrix is equal to LU =  −2 1 0   0 2 −4 .
0 4 1 0 0 −1
First we solve LY = B for Y , we have
    
1 0 0 y1 0
 −2 1 0   y2  =  4  .
0 4 1 y3 17
The top line implies that y1 = 0. The middle line states that −2y1 + y2 = 4 and therefore y2 = 4.
The last line tells us that 4y2 + y3 = 17 and therefore y3 = 1.
Finally we solve U X = Y for X, we have
    
3 1 6 x1 0
 0 2 −4   x2  =  4  .
0 0 −1 x3 1
The bottom line shows that x3 = −1. The middle line  then shows
 that x2 = 0, and then the top
2
line gives us that x1 = 2. The required solution is X =  0 .
−1

3. Do matrices always have an LU decomposition?

No. Sometimes it is impossible to write a matrix in the form “lower triangular”דupper triangular”.

Why not?
An invertible matrix A has an LU decomposition provided that all its leading submatrices have
non-zero determinants. The k th leading submatrix of A is denoted Ak and is the k × k matrix found
by looking only at the top k rows and leftmost k columns. For example if
 
1 2 4
A =  3 8 14 
2 6 13
then the leading submatrices are
 
  1 2 4
1 2
A1 = 1, A2 = , A3 =  3 8 14  .
3 8
2 6 13
The fact that this matrix A has an LU decomposition can be guaranteed in advance because none
of these determinants is zero:
|A1 | = 1,
|A2 | = (1 × 8) − (2 × 3) = 2,
8 14 3 14 3 8
|A3 | = −2 +4 = 20 − (2 × 11) + (4 × 2) = 6
6 13 2 13 2 6
(where the 3 × 3 determinant was found by expanding along the top row).

Decomposition
 72

Example 7 
1 2 3
Show that  2 4 5  does not have an LU decomposition.
1 3 4

Solution
The second leading submatrix has determinant equal to
1 2
= (1 × 4) − (2 × 2) = 0
2 4
which means that an LU decomposition is not possible in this case.

Task
Which, if any, of these matrices have an LU decomposition?
 
    1 −3 7
3 2 0 1
(a) A = , (b) A = , (c) A =  −2 6 1 .
0 1 3 2
0 3 −2

Your solution
(a)

Answer
|A1 | = 3 and |A2 | = |A| = 3. Neither of these is zero, so A does have an LU decomposition.

Your solution
(b)

Answer
|A1 | = 0 so A does not have an LU decomposition.

Your solution
(c)

Answer
|A1 | = 1, |A2 | = 6 − 6 = 0, so A does not have an LU decomposition.
 73 ®

Can we get around this problem?

Yes. It is always possible to re-order the rows of an invertible matrix so that all of the submatrices
have non-zero determinants.

Example 8  
1 2 3
Reorder the rows of A =  2 4 5  so that the reordered matrix has an LU
1 3 4
decomposition.

Solution
Swapping the first and second rows does not help us since the second leading submatrix will still
have a zero determinant. Let us swap the second and third rows and consider
 
1 2 3
B= 1 3 4 
2 4 5
the leading submatrices are
 
1 2
B1 = 1, B2 = , B3 = B.
1 3
Now |B1 | = 1, |B2 | = 3 × 1 − 2 × 1 = 1 and (expanding along the first row)
|B3 | = 1(15 − 16) − 2(5 − 8) + 3(4 − 6) = −1 + 6 − 6 = −1.
All three of these determinants are non-zero and we conclude that B does have an LU decomposition.

 
Task 1 −3 7
Reorder the rows of A =  −2 6 1  so that the reordered matrix has an
0 3 −2
LU decomposition.

Your solution

HELM (2008): 29
 74

Answer
Let us swap the second and third rows and consider
 
1 −3 7
B= 0 3 −2 
−2 6 1
the leading submatrices are
 
1 −3
B1 = 1, B2 = , B3 = B
0 3
which have determinants 1, 3 and 45 respectively. All of these are non-zero and we conclude that
B does indeed have an LU decomposition.

Exercises
1. Calculate LU decompositions for each of these matrices
   
  2 1 −4 1 3 2
2 1
(a) A = (b) A =  2 2 −2  (c) A =  2 8 5 
−4 −6
6 3 −11 1 11 4
2. Check each answer in Question 1, by multiplying out LU to show that the product equals A.

3. Using the answers obtained in Question 1, solve the following systems of equations.
    
2 1 x1 1
(a) =
−4 −6 x2 2
    
2 1 −4 x1 4
(b)  2 2 −2   x2  =  0 
6 3 −11 x3 11
    
1 3 2 x1 2
(c)  2 8 5   x2  =  3 
1 11 4 x3 0
 
1 6 2
4. Consider A =  2 12 5 
−1 −3 −1

(a) Show that A does not have an LU decomposition.

(b) Re-order the rows of A and find an LU decomposition of the new matrix.
(c) Hence solve

x1 + 6x2 + 2x3 = 9
2x1 + 12x2 + 5x3 = −4
−x1 − 3x2 − x3 = 17

s
 75 ®

Answers

1. (a) We let
      
2 1 1 0 U11 U12 U11 U12
= LU = = .
−4 −6 L21 1 0 U22 L21 U11 L21 U12 + U22
Comparing the left-hand and right-hand sides row by row gives us that U11 = 2, U12 = 1,
L21 U11 = −4 which implies that L21 = −2 and, finally, L21 U12 + U22 = −6 from which
we see that U22 = −4. Hence
    
2 1 1 0 2 1
=
−4 −6 −2 1 0 −4
is an LU decomposition of the given matrix.
(b) We let
   
2 1 −4 U11 U12 U13
 2 2 −2  = LU =  L21 U11 L21 U12 + U22 L21 U13 + U23  .
6 3 −11 L31 U11 L31 U12 + L32 U22 L31 U13 + L32 U23 + U33
Looking at the top row we see that U11 = 2, U12 = 1 and U13 = −4. Now, from the
second row, L21 = 1, U22 = 1 and U23 = 2. The last three unknowns come from the
bottom row: L31 = 3, L32 = 0 and U33 = 1. Hence
    
2 1 −4 1 0 0 2 1 −4
 2 2 −2  =  1 1 0   0 1 2 
6 3 −11 3 0 1 0 0 1
is an LU decomposition of the given matrix.
(c) We let
   
1 3 2 U11 U12 U13
 2 8 5  = LU =  L21 U11 L21 U12 + U22 L21 U13 + U23  .
1 11 4 L31 U11 L31 U12 + L32 U22 L31 U13 + L32 U23 + U33
Looking at the top row we see that U11 = 1, U12 = 3 and U13 = 2. Now, from the
second row, L21 = 2, U22 = 2 and U23 = 1. The last three unknowns come from the
bottom row: L31 = 1, L32 = 4 and U33 = −2. Hence
    
1 3 2 1 0 0 1 3 2
 2 8 5  =  2 1 0  0 2 1 
1 11 4 1 4 1 0 0 −2
is an LU decomposition of the given matrix.

2. Direct multiplication provides the necessary check.

 76
Answers
3.

(a) We begin by solving

    
1 0 y1 1
=
−2 1 y2 2
Clearly y1 = 1 and therefore y2 = 4. The values y1 and y2 appear on the right-hand side
of the second system we need to solve:
    
2 1 x1 1
=
0 −4 x2 4
The second equation implies that x2 = −1 and therefore, from the first equation, x1 = 1.
(b) We begin by solving the system
    
1 0 0 y1 4
 1 1 0  y2  =  0  .
3 0 1 y3 11
Starting with the top equation we see that y1 = 4. The second equation then implies
that y2 = −4 and then, from the third equation, y3 = −1. These values now appear on
the right-hand side of the second system
    
2 1 −4 x1 4
 0 1 2   x2  =  −4  .
0 0 1 x3 −1
The bottom equation shows us that x3 = −1. Moving up to the middle equation we
obtain x2 = −2. The top equation yields x1 = 1.
(c) We begin by solving the system
    
1 0 0 y1 2
 2 1 0  y2 = 3  .
 
1 4 1 y3 0
Starting with the top equation we see that y1 = 2. The second equation then implies
that y2 = −1 and then, from the third equation, y3 = 2. These values now appear on
the right-hand side of the second system
    
1 3 2 x1 2
 0 2 1   x2  =  −1  .
0 0 −2 x3 2
The bottom equation shows us that x3 = −1. Moving up to the middle equation we
obtain x2 = 0. The top equation yields x1 = 4.
 77®
Answers
4.

(a) The second leading submatrix has determinant 1 × 12 − 6 × 2 = 0 and this implies that
A has no LU decomposition.
 
1 6 2
(b) Swapping the second and third rows gives  −1 −3 −1  . We let
2 12 5
   
1 6 2 U11 U12 U13
 −1 −3 −1  = LU =  L21 U11 L21 U12 + U22 L21 U13 + U23  .
2 12 5 L31 U11 L31 U12 + L32 U22 L31 U13 + L32 U23 + U33
Looking at the top row we see that U11 = 1, U12 = 6 and U13 = 2. Now, from the
second row, L21 = −1, U22 = 3 and U23 = 1. The last three unknowns come from the
bottom row: L31 = 2, L32 = 0 and U33 = 1. Hence
    
1 6 2 1 0 0 1 6 2
 −1 −3 −1  =  −1 1 0   0 3 1 
2 12 5 2 0 1 0 0 1
is an LU decomposition of the given matrix.
(c) We begin by solving the system
    
1 0 0 y1 9
 −1 1 0   y2  =  17  .
2 0 1 y3 −4

(Note that the second and third rows of the right-hand side vector have been swapped
too.) Starting with the top equation we see that y1 = 9. The second equation then
implies that y2 = 26 and then, from the third equation, y3 = −22. These values now
appear on the right-hand side of the second system
    
1 6 2 x1 9
 0 3 1   x2  =  26  .
0 0 1 x3 −22
The bottom equation shows us that x3 = −22. Moving up to the middle equation we
obtain x2 = 16. The top equation yields x1 = −43.
 78

SECTION-3.
RANK OF A MATRIX.

3.1 Sub Matrix A matrix obtained from a matrix A by omitting some of its rows or
columns
(or both) is called a sub matrix of A
For Example the matrices
2 1 9
1 9 5
2 1 2 1 9
= , = , = 4 1 0 , = 3 4 1 etc. are the sub
3 4 3 4 1 5 7 2
7 2 1 1 7
4
2 1 9 5
matrices of a matrix = 3 4 1 0 in which and are square sur matrices of of
5 7 2 1
4 1 7 7
order 2 and 3 respectively of A .
As in care of set, every matrix is considers as in sub matrix of itself

3.2 Minor of A Matrix The determinant of a square sub matrix of order r is called a minor
of order r of A. A minor of order r is said to be r-rowed minor.
2 1 1 2
For Example-Consider a matrix = 0 2 0 4
3 3 2 6
2 1 1
On omitting forth column , we get a square sub matrix = 0 2 0 and its determinant
3 3 2
2 1 1
i.e.| |= 0 2 0 = 2 is a minor of order 3
3 3 2
On omitting third columns we get square matrix
2 1 2 2 1 2
= 0 2 4 and | |= 0 2 4 is also a minor of order 3
3 3 6 3 3 6
On omitting third row and third & forth columns , we get the square sub matrix =
2 1
and | | is minor of order 2 .
0 2
1 2
On omitting second row and second & third columns we get square matrix =
3 6
and | | is again a minor of order 2 since | | = 2 ≠ 0 &| | = 4 ≠ 0, then these are
 79
called non zero (or non vanishing ) minors whereas | | = 0 &| | = 0, so these are
called zero (or vanishing ) minors of A

Remarks (a) from this Example it is also clear that there may be more than one minors
of the same order obtained by deleting different rows and\or columns

(b) The order of the largest order minor of a matrix of order × =

min.( , )
5 2 3 1
For Example the order of the largest order minor of a matrix 4 6 2 8 is 3
9 5 6 2 ×

and (c) the order of the largest order minor of a n-square matrix = n

2 3 1 5
e.g. the order of the largest order minor of 1 0 3 1 is 4 .
2 6 4 4
5 1 0 2

3.3 RANK OF A MATRIX

The rank of a matrix A is r (a positive integer) if at least one minor of order r of A is non
zero whereas it’s all minors of order ( + 1) zero. The rank of a matrix A denoted by
( ) or ( ) .
Important Facts
i. For a non zero matrix A, ( ) ≥ 1
ii. The rank of a null matrix of any order is zero
iii. The rank of an × matrix A is less than or equal to the min. {m,n}
i,e. ( ) ≤ min. { , }
iv. The rank of a non- singular square matrix is always equal to its order.
v. ( ) = , where In is a unit matrix of order
vi. ( )= ( )=
vii. The rank of the product of two matrices cannot exceed the rank of either matrix i.e.
( ) ≤ ( ) or ( )
viii. The rank of a matrix remains unchanged by applying elementary transformations i.e.
the equivalent matrices have the same rank
ix. Let A is a non zero column matrix and B is a non zero row matrix then ( )=1.
x. If A and B be two equivalent matrices, then rank A = rank B.
 80
xi. If A and B are equivalent matrices, then there exist non-singular matrices P
and Q such that B = PAQ.
xii. If two matrices A and B have the same size and the same rank, they are
equivalent.
xiii. Two m x n matrices are equivalent if and only if they have the same rank.

3.3.1 Nullity of a matrix- If A is a square matrix of order then − ( ) is called the

nullity of the matrix A and is denoted by ( ). Thus the nullity of a non-singular square
matrix of order is zero.

3.4 METHODS TO FIND RANK OF A MATRIX

The methods, which are often used of find rank of the matrices, can be described as

1. By using definition with or without means of the elementary transformations

2. By reducing into normal form
3. By reducing into echelon form.

1. By Using Definitions –
From definition of rank, the rank of a matrix A is the order of the largest order non zero
minor of A so that to find the rank of a matrix A, we have to identify the order of the largest
non- zero minor .But in the case of matrices of large order, this process, involves a lot of
computations, So it is tedious.
To reduce computations, we transform maximum possible entries of A to zero by applying
elementary transformations and then use definition.

SOLVED EXAMPLES
5 −1 0
Example 1 Find the rank of matrix = 3 1 2
2 3 6

Solution- Since A is a square matrix then we have

5 −1 0
| | = 3 1 2 = 5(6 − 6) + 1(18 − 4) + 0. (9 − 2)
2 3 6
 81
= 0 + 14 + 0

= 14 ≠ 0

i.e.| | ≠ 0 which is the largest non zero minor of A therefore( ) = 3

Example 2 Find the rank of the following matrices

1 2 3
i. = 3 4 5
4 5 6

0 −
ii. = − 0
− 0

0 2 3
iii. = 0 4 6
0 6 9
1 2 3
Solution (i) we have | | = 3 4 5
4 5 6

= 1(24 − 25) − 2(18 − 20) + 3(15 − 16)

= −1 + 4 − 3 = 0 | | = 0 i.e, minor

| | = 0 i.e, minor of order 3 is zero

⇒ ( )<3

Further the minor of order 2 obtained by deleting third row & third column

1 2
= 4 − 6 = −2 ≠ 0
3 4

Hence ( ) = 2

0 −
(ii) We have | | = − 0 = 0(0 + ) − (0 − ) − ( − 0)
− 0
=0+ − =0

∴ ( )<3
 82
And the minor of 2 obtained by deleting the first row & third column

− 0
= − 0 = −1 ≠ 0
−

Hence ( )=2

(iii) It is obvious that | | = 0 and also all minors of order 2 are zero but each element of
C is not zero, therefore ( )=1

1 2 3 −4
Example 3 Find the rank of matrix : A = −2 3 7 −1
1 9 16 −13

Solution

1 2 3 −4
A = −2 3 7 −1
1 9 16 −13

→ +2 , → −

1 2 3 −4
A~ 0 7 13 −9
0 7 13 −9

→ −

1 2 3 −4
A~ 0 7 13 −9
0 0 0 0

Here the number of non-zero rows is 2 , therefore Rank (A) = 2 .

 1 2 3 4
Example 4 Find the rank of matrix : A =  2 3 7  1 
 
 1 9 16  13

 1 2 3 4
Solution: A =  2 3 7  1 
 
 1 9 16  13

→ +2 , → −
 83
1 2 3  4
A ~ 0 7 13  9 
 
0 7 13  9 

→ −

1 2 3  4 
A ~ 0 7 13  9 
 
0 0 0 0 

Here the number of non-zero rows is 2 , therefore Rank (A) = 2 .

2. By Reducing Into Normal Form

Any non-zero matrix A of rank r can be reduced by a sequence of elementary

0
transformations to the Form , where is a unit matrix of order r. This form is called
0 0
a normal form of A.

Other normal forms are , , [ , 0].

0

Theorem 1 Let A be an m × n matrix of rank r. Then there exist non-singular matrices P

0
and Q of orders m and n respectively such that PAQ =
0 0

NOTE-

(i) Each elementary row transformation of A is equivalent to pre multiplying A by the

corresponding elementary matrix.

(ii) Each elementary column transformation is equivalent to post multiplying A by the

corresponding elementary matrix.

So, there exist elementary matrices , ,…, and , … such that

0
, ,…, A , … =
0 0

0
=
0 0
 84
where = , ,…, & = , … ,

Working Rule to Find Normal Form and Non Singular Matrices P&Q-

Let A be a non-zero m × n matrix.

Write = (which is obviously true).

Reduce A on the L. H. S to normal form by applying elementary row and column

transformations on A in such a way that the each elementary row transformation applied on
A will be also applied to on R. H. S and each elementary column transformation applied
on A will be applied to on R. H. S.

After a sequence of suitable applications of elementary transformations, we get

0
=
0 0

Then the rank of A = the rank of =

SOLVED EXAMPLES

0 1 2 1
Example 1 Reduce the matrix 1 2 3 1 to normal form and hence find the rank.
3 1 1 3

0 1 2 1 1 0 2 1
Solution Let = 1 2 3 1 ~ 2 1 3 1 ↔
3 1 1 3 1 3 1 3

1 0 0 0 → + (−2)
~ 2 1 −1 1 → −
1 3 −1 2

1 0 0 0 → + (−2)
~ 0 1 0 0
→ −
0 3 2 2

1 0 0 0
~ 0 1 0 0 → − (−3)
0 0 2 2

1 0 0 0 1
~ 0 1 0 0 →
2
0 0 1 1
 85
1 0 0 0
~ 0 1 0 0 → −
0 0 1 0

= [ : 0]

This is the normal form of A and so (A) = 3

Example 2 Find two non-singular matrices P and Q such that PAQ is in the normal form

1 −1 −1
Where = 1 1 1 . Also find rank of A.
3 1 1

1 −1 −1
Solution Given = 1 1 1
3 1 1 ×

We have =

1 −1 −1 1 0 0 1 0 0
1 1 1 = 0 1 0 0 1 0
3 1 1 0 0 1 0 0 1

To obtain matrices P & Q, we reduce the LHS matrix to normal form under the fact that

row operations to be applied to pre factor and column operations to be applied to post factor.

Apply, → +

→ +

1 0 0 1 0 0 1 1 1
1 2 2 = 0 1 0 0 1 0
3 4 4 0 0 1 0 0 1

and to post factor

→ − 1 0 0 1 0 0 1 1 1
0 2 2 = −1 1 0 0 1 0
→ + (−3)
0 4 4 −3 0 1 0 0 1

And to pre factor

 86
1 0 0
1 1 0 0 1 1 1 1 1
→ 0 1 1 = − 0 0 1 0
2 2 2
0 0 0 0 0 1
−1 −2 1

And to pre factor

1 0 0
1 0 0 1 1 1 1 0
→ − 0 1 0 = − 0 0 1 −1
0 0 1 2 2 0 0 1
−1 −2 1

And to post factor

0
=
0 0

It is the normal form of the given matrix and hence the rank of A i.e. (A) = 2and

1 0 0
1 1 1 1 0
= − 0 , = 0 1 −1
2 2 0 0 1
−1 −2 1

1 −1 −1 2
Example 3 Let = 4 2 2 −1 . Find the non-singular matrices P and Q, such that
2 2 0 −2
PAQ is in the normal form. Also find the rank of A.

1 −1 −1 2
Solution Given = 4 2 2 −1
2 2 0 −2 ×

We have =

1 0 0 0
1 −1 −1 2 1 0 0
0 1 0 0
4 2 2 −1 = 0 1 0
0 0 1 0
2 2 0 −2 0 0 1
0 0 0 1

To obtain matrices P & Q, we reduce the LHS matrix to normal form under the fact that

row operations to be applied to pre factor and column operations to be applied to post factor.
 87
→ + (−4) 1 1 1 −2
1 0 0 0 1 0 0
→ + (−2) 0 1 0 0
0 6 6 −9 = −4 1 0
0 0 1 0
0 4 2 −6 −2 0 1
0 0 0 1

1
1 0 0 1 1 1 −
3 1 0 0 0 4 1 2
→ + − 0 3
2 0 2 2 0 = 3 3 0 1 0
0 2 1 0 1 2
−1 0 0 0 1 0
2
0 0 0 1

1
1 0 0 1 1 0 −
1 0 0 0 2 1 2
→ − − 0 3
0 2 0 0 = 3 6 0 1 −1
0 0 −1 0 1 1 1 2
− 0 0 1 0
3 3 2
0 0 0 1

[ : 0] = ,

1 0 0 1 1 0 −

Where = − 0 , = 0 1 −1
− 0 0 1 0
0 0 0 1

and the rank of A = 3

Example 4 Find non-singular matrices P , Q so that PAQ is a normal form where

 2 1  3  6
A = 3  3 1 2 
1 1 1 2 

Solution Order of A is 3 × 4

Total number of rows in A = 3 Hence consider unit matrix ɪ

Total number of column in A = 4

Hence consider unit matrix ɪ

× =ɪ ɪ
 88
1 0 0 0
 2 1  3  6 0 0 0
3  3 1 
1 1 0 0
 2 = 0 1 0 A
 0 1 0
1 1 1 2  0 0 1 0 
0 0 0 1

1 0 0 0
1 1 1 2 0 1 0
3  3 1  =
0 1 0 0
 2  0 1 0 A
 0 1 0
2 1  3  6 1 0 0 0 
0 0 0 1

→ −3 , → −2

1 0 0 0
1 1 1 2  0
0  6  2  4  =
0 0 1 1 0 0
  0 1 −3 A 
 0 1 0
0  1  5  10 1 0 −2 0 
0 0 0 1

→ − , → − , → −2

1  1  1  2
1 0 0 0  0 1 0
0  6  2  4  =
0 0 1 0 
  0 1 −3 A 
 0
0  1  5  10 1 0 −2 0 0 1 
0 0 0 1

→− , →− , ↔

1  1  1  2
1 0 0 0  0 1 0 1 0
0 1 5 10  =
0 0 
  −1 0 2 A
 0
0 6  2  4 0 −1 3 0 0 1 
0 0 0 1

→ −6

1  1  1  2
1 0 0 0  0 1 0
0 1  =
0 0 1
 0 
 5 10  −1 0 2 A
 0
6 −1 −9 0 0 1
0 0  28  56 
0 0 0 1
 89
→ −5 , → − 10

1  1 4 8 
1 0 0 0  0 0 1 0 1  5  10
0 1  =
 0 0  −1 0 2 A 
0 0 1 0 
0 0  28  56 6 −1 −9  
0 0 0 1 

1
→−
28

1  1 4 8 
1 0 0 0  0 0 1  
0 1 0 0 = −1 0 2 A 0 1  5  10
  0 0 1 0 
0 0 1 2 −
 
0 0 0 1 

→ −2

1  1 4 0 
0 1  5 0 
 
0 0 1  2
1 0 0 0  0 0 1  
0 1 0 0 = −1 0 2 A 0 0 0 1 
 
0 0 1 0 −

N = PAQ where

1  1 4 0 
0 1  5 0 
 
0 0 1  2
0 0 1  
P = −1 0 2 and Q = 0 0 0 1 
−

Note: P and Q are not unique.

3. By Reducing Into Echelon Form

A matrix is said to be in Echelon form if

(i) the number of zeros preceding the first non-zero element in each row is less than that
in the subsequent row(s),
 90
(ii) all zero rows (i.e., rows with zero elements only), if any, are on the bottom of the
matrix i.e. all the non- zero rows precede the zero rows,

(iii) the first non-zero element in each row is unity.

2 1 −1 0
1 2 3
For Example- = 0 1 2 , = 0 0 −1 2
0 0 0 0
0 0 0
0 0 0 0

1 −1 0 4 5 1 2 3
= 0 −1 2 1 3 = 0 2 1 are in Echelon forms
0 0 0 6 1
0 0 2
0 0 0 0 0

Result: If a matrix A is reduced to Echelon form then rank of A i.e. (A) = the number of
non-zero rows in its Echelon form

Thus in the above Examples, (A) = 2, (B) = 2, (C) = 3 and (D) = 3.

SOLVED EXAMPLES

1 2 3 0
Example 1 Find the rank of the matrix = 2 4 3 2 , by reducing to Echelon form.
3 2 1 3
6 8 7 5

Solution

1 2 3 0 1 2 3 0 → + (−2)
Given = 2 4 3 2 ~ 0 0 −3 2 → + (−3)
3 2 1 3 0 −4 −8 3
6 8 7 5 0 −4 −11 5 → + (−6)

1 2 3 0
~ 0 0 −3 2 → −
0 −4 −8 3
0 0 −3 2

1 2 3 0
~ 0 −4 −8 3 ↔
0 0 −3 2
0 0 −3 2
 91
1 2 3 0
~ 0 −4 −8 3 →
0 0 −3 2
0 0 −3 2

1 2 3 0
~ 0 −4 −8 3 → −
0 0 −3 2
0 0 0 2

which is echelon form

∴ (A) = the number of non-zero rows in the Echelon form = 3

1 2 3 −1
Example 2 Determine the rank of the matrix = 3 6 9 −3 , by reducing to echelon
2 4 6 −2
form.

Solution

1 2 3 −1 1 2 3 −1 → + (−3)
Given = 3 6 9 −3 ~ 0 0 0 0 → + (−2)
2 4 6 −2 0 0 0 0

= B, which is a row echelon matrix.

∴ r(A) = the number of non-zero rows in B = 1

6 3 5 9
Example 3 Find the value of k if the rank of the matrix 5 2 3 6 is 3.
3 1 2 3
2 1 1

6 3 5 9 1
Solution = 5 2 3 6 ~ 5 2 3 6 →
3 1 2 3 3 1 2 3
2 1 1 2 1 1
 92
1 5 3
1
2 6 2
1 3 3
0 − − − → + (−5)
~ 2 6 2 → + (−3)
3
1 − 3 → + (−2)
0 − 6 −
0 2 4 2
0 − −3
6

1 5 3
1
2 6 2
1 3 3
0 − − −
~ 2 6 2 → −
4
0 0 6 0
0 0 4 −3
−
6

1 5 3
1
2 6 2
1 3 3
~ 0 −2 −
6
−
2 → +
4
0 0 0
0 0 6 −3
0

=B

Given r(A) = 3. So, the number of non-zero rows of B should be 3.

∴k−3=0⇒k=3
 93
EXERCISE

Find the rank of the following matrices reducing to echelon form

2 3 −1 −1
1. 1 −1 −2 −4
3 1 3 −2
6 3 0 −7
0 1 −3 −1
2. 1 0 1 1
3 1 0 2
1 1 −2 0
1 2 3 0
3. 2 4 3 2
3 2 1 3
6 8 7 5
4 3 0 2
4. 3 4 −1 3
−7 −7 1 5
1 2 −1 3
5. 4 1 2 1
3 −1 1 2
1 2 0 1
1 1 2 1
−1 −1 −2 1
6. 1 2 1 −1
1 3 0 −3
1 1 2 3
1 1 1 1
7. 1 3 −2 1
2 0 −3 2
3 3 −3 3
1 2 −2 3
8. 2 5 −4 6
−1 −3 2 −2
2 4 −4 6
3 1 −5 −1
9. 1 −2 1 −5
1 5 −7 2
1 3 4 3
10. Find the rank of the matrix A = 3 9 12 3 , by reducing to an echelon matrix
1 3 4 1
−2 −1 −1
11. Find the rank of the matrix 12 8 6
10 5 6
 94
2 1 −3 −6
12. Reduce the matrix 3 −3 1 2 to normal form and hence find the rank
1 1 1 2
4 4 −3 1
13. Find the values of k if the rank of 1 1 −1 0 is 3.
2 2 2
9 9 3
2 1 −1 3
14. Find the values of a and b if the matrix 1−1 2 4 is of rank 2.
7−1
1−2 3 1
15. Find the values of a and b if the matrix 2 1 −1 2 is of rank 2.
6−2
1 −1 2 −3
16. Reduce to normal form and find the rank of 4 1 0 2 .
0 3 1 4
0 1 0 2
0 1 2 1
17. Reduce to normal form and find the rank of 1 2 3 2 .
3 1 1 3
1 1 2
18. If = 1 2 3 then find non-singular matrices P and Q such that PAQ is in
0 −1 −1
normal form and find its rank.
19. If A is a non-zero 3 × 1 matrix and B is a non-zero 1 × 3matrix , then the rank of
AB is:
(a) 0
(b) 1
(c) 2
(d) 3

20. Matrix

1 0 0 0 1
0 1 0 0 1 

0 0 0 1 1
 
0 0 0 0 0

is :
 95

(a) An identity matrix (b)A scalar matrix (c)In normal form (d)In echelon

form

Answers

1. 3
2. 2
3. 3
4. 2
5. 3
6. 3
7. 3
8. 3
9. 3
10. 2
11. 3
12. 3
13. 2
14. 4,18
15. 4,6
16. [ , 0], =4
17. [ , 0], =3
1 0 0 1 −1 −1
18. = −1 1 0 , = 0 1 −1 and rank = 2
−1 1 1 0 0 1
19. (b)
20. (d)