Chapter Two

2. Matrices and Determinants

INTRODUCTION: Information in science and mathematics is often organized into rows and
columns to form rectangular arrays, called “matrices” (plural of “matrix”). Matrices are often
tables of numerical data that arise from physical observations, but they also occur in various
mathematical contexts. For example, we shall see in this chapter that to solve a system of

` + 2a = 5
equations such as

3` − a = 1

1 2 5
all of the information required for the solution is embodied in the matrix

é ê
3 −1 1
and that the solution can be obtained by performing appropriate operations on this matrix. This is
particularly important in developing computer programs to solve systems of linear equations
because computers are well suited for manipulating arrays of numerical information. However,
matrices are not simply a notational tool for solving systems of equations; they can be viewed as
mathematical objects in their own right, and there is a rich and important theory associated with
them that has a wide variety of applications. In this chapter we will begin the study of matrices.
2.1. Definition of Matrix
Definition:-A matrix is a rectangular array of numbers or variables, which we will enclose in
brackets. The numbers (or variables) are called entries or, less commonly, elements of the

columns. A matrix with ë rows and H columns has the form:

matrix. The horizontal lines of entries are called rows, and the vertical lines of entries are called

OPP OP> ⋯ OPA

O>P O>> ⋯ O>A
­ =ì ⋮ ⋮ ⋱ ⋮ ðOr­ = ñOœò ó, where € = 1, 2, ..., ç and ˜ = 1,2, … , d
OïP Oï> ⋯ OïA
By an ë × Hmatrix (read as“ë by H matrix”) we mean a matrix with ërows and Hcolumns—
rows always come first! is called the size/order/shape/dimension of the matrix. We shall denote
matrices by capital boldface letters A, B, C, …, or by writing the general entry in brackets; like
ô = ñG“¥ ó, and so on..The element G“¥ ,is called the “¥ entry, appears in row “ and column ¥.Thus
GqU is the entry in Row 2 and Column 1.

27
 1 4 −12 4 6 9
Example: Consider the following matrices

­ = õ 5 −1 10 9ö æ = õ6 0 −2ö
−1 −8 7 5 5 −8 1
The dimension/size of matrix A is 3×4 and the dimension/size of matrix B is 3×3. The entry O@@
in matrix A is 7 and the entry O>p in matrix A is 9.
NB: Matrices are important because they let us to express large amounts of data and functions in

3 4 2 −3
an organized and concise form.

Example: Given a matrix A = õ 1 −1 10 9 ö, then

−10 −8 7 5
a) Find the size of A c) List rows of A
b) List columns of A d) List elements of A

1. Let ‚ be a real number. Assume that B =£‚¤. Then, determine:

Exercise:

a) The size of B c) the no of columns of B

b) The rows of B d) the elements B
2. Construct a 3x3 matrix whose OϘ entry is given by 2j Рi

2.2. Types of Matrices

In matrix theory, there are many special kinds of matrices that are important because they
possess certain properties. The following is a list of some of these matrices.
i. Zero Matrix (Null Matrix): Matrix that consists of all zero entries is called a zero
matrix and is denoted by bold zero,T orTë¹H .
0 0 0
Example: 0@Â@ = õ0 0 0ö
0 0 0
Square Matrix: any ç`d matrix is called square matrix if ç = d. The order of d`d
square matrix is d`d or simply d.
ii.

1 5 −3
Example:­@Â@ = õ2 0 −2ö
2 3 7
iii. Rectangular Matrix: A matrix of any size is called a rectangular matrix; this includes

Row Matrix (row vector): A 1 × d matrix is called a row matrix (row vector).
square matrices as a special case.
iv.

28
 ­ = £OPP OP> … OPA ¤ = (OP , O> , … , OA )
Column Matrix (column vector):A d`1 matrix is called a column matrix (column
O1
v.

O
ù 2ü
­=ø ⋮ û
⋮
vector)

÷Od ú
vi. Diagonal Matrix: A d × dsquare matrix ýis said to be a diagonal matrix if all its

ePP 0 … 0
entries except the main diagonal entries are zeros.

0 e>> … 0
ý=ì ⋱ ð
⋮ ⋮ ⋮
0 0 … eAA
Remark: A diagonal matrix each of whose diagonal elements are equal is called a scalar
matrix.

1 0 … 0
vii. Identity Matrix (Unit Matrix): is a diagonal matrix in which all diagonal elements are 1.

0 1 … 0
þA = ì ⋱ ⋮ ð is an d × d identity matrix.
⋮ ⋮
0 0 … 1
viii. Triangular Matrix: 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

OPP OP> … OPA OPP 0 … 0

or lower triangular is called triangular.

0 O>> … O O O>> … 0
­=ì ⋮ ð æ=ì ⋮ ð
>A >P
⋮ ⋮ ⋱ ⋮ ⋱ ⋮
0 0 … OAA OAP OA> … OAA

per Triangular Lower Triangular

Remark:Observe that diagonal matrices are both upper triangular and lower triangular since

q T T q −‘ U
they have zeros below and above the main diagonal.

Example:ô = õ† / Tö
= õT / ‘ö
U U −q T T −q

Lower triangular matrix Upper triangular matrix

29
 2.3. Basic Operations on Matrices

Definition: Two matrices are defined to be equal, denoted by  = ,if they have the same size
Equality of Matrices

Example: Find the value of ` and a if matrix ­ is equal to matrixæ.

and their corresponding entries are equal.

`+1 3 4 3
­=é ê and æ =  
0 6 0 a−2
Solution: the two matrices are equal if and only if ` = 3 anda = 8.

Definition: Let A = ñ Oœò óand B = ñ gœò óbe two matrices of the same size, say, ç × d matrices.
Matrix Addition and Subtraction

Then the sum of A and B, written A + B, is the matrix obtained by adding the entries of B to the
corresponding entries of A. The difference A – B is the matrix obtained by subtracting the entries
of B from the corresponding entries of A. In matrix notation, if A = ñ Oœò óand B = ñ gœò óhave the
same size, then
­ + æ = £O¤œò + £g¤œò = £Oœò + gœò ¤ Ode ­ − æ = £Oœò ¤ï×A − £gœò ¤ï×A = £Oœò − gœò ¤ï×A

−1 2 3 4 1 6 −1 3
Example: Consider the following matrices

­=õ 1 0 2 5ö , æ = õ2 5 0 −4ö
9 −2 0 6 3 4 1 2
Then find­ + æ Ode ­ − æ
−1 + 1 2 + 6 3 − 1 4 + 3 0 8 2 7
Solution: ­ + æ = õ 1 + 2 0 + 5 2 + 0 5 − 4ö = õ 3 5 2 1ö
9 + 3 −2 + 4 0 + 1 6 + 2 12 2 1 8
−1 − 1 2 − 6 3 + 1 4 − 3 −2 −4 4 1
­−æ =õ 1−2 0 − 5 2 − 0 5 + 4ö = õ−1 −5 2 9ö
9 − 3 −2 − 4 0 − 1 6 − 2 6 −6 −1 4
Remark: Matrices of different sizes cannot be added or subtracted.

Scalar Multiplication of Matrix

Definition: If A is any matrix and c is any scalar; then the product cA is the matrix obtained by

OPP OP> … OPA

multiplying each entry of the matrix A by c. The matrix cA is said to be a scalar multiple of A.i.e.

O>P O>> … OPA

—­ = — ì ⋮ ⋮ ⋱ ⋮ ð
OïP Oï> … OïA

30
 —OPP —OP> … —OPA
—O>P —O>> … —OPA
=ì ⋮ ⋮ ⋱ ⋮ ð = ñ—Oœò óïÂA
—OïP —Oï> … —OïA
−1 2 3 4 1 6 −1 3
Example: Let­ = õ 1 0 2 5ö and æ = õ2 5 0 −4ö. Then compute the following
9 −2 0 6 3 4 1 2
a. 2­ − æ P
­ + 2­
@
b.
−2 4 6 8 1 6 −1 3 −3 −2 7 5
Solution: a. 2A − B = õ 2 0 4 10ö − õ2 5 0 −4ö = õ 0 −5 4 14ö
18 −4 0 12 3 4 1 2 15 −8 −1 10
1 7
YP > p YZ Pp >W
@ @ @ −2 4 6 8 @ @ @
P Z
b. A + 2A =  @ 0 +õ 2 0 4 10ö =  @ 0
P > X Pp @X

 
@ @ @ @ @
18 −4 0 12
3 0 2 21 0 14
Y> YPp
@ @

1. Find the values of ` and afor the following matrix equation.

Exercise:

`+2 a+3 3 a
2é ê=é ê
3 0 6 b
2 −1
2. Find matrix ­ if 2­ = é ê
6 0
Remark: If ­P , ­> , …,­A are matrices of the same size and —P, —> , …,—A are scalars, then an
expression of the form—P ­P + —> ­> + ⋯ + —A ­A is called a linear combination of ­P , ­> , …,­A
with coefficients—P, —> , …,—A .

Suppose A, B, and C are ç × dmatrices (having the same size) and j and ” are scalars. Then
Properties of Matrix Addition and Scalar Multiplication

i. ­ + æ = æ + ­(commutative law of addition)

ii. (­ + æ) + = ­ + (æ + )(Associative law of addition)
iii. ­ + T = T + ­ (Existence of additive identity)
iv. ­ + (−­) = (−­) + ­ = T(Existence of additive inverse)
v. j(­ + æ) = j­ + jæ
vi. (j + ”)­ = j­ + ”­
vii. (j( ”­)) = j( ”­)
viii. 1­ = ­ and 0­ = T

31
 2.4. Product of Matrices
Definition: Let matrix ­ be an ç × d matrix and æ be an d × ™ matrix (i.e. the number of
columns of ­ is equal to the number of rows of æ). Then the product of ­ and æ, denoted by ­æ,
is an ç × ™ matrix which is obtained by multiplying the corresponding elements of row € of ­ by
column ˜ of æ and adding the product, i.e. if ­ = ñOœò óï×A and æ = ñgò óA× , then ­æ = =

£—œ ¤ï× , where —œ = AòP Oœò gò , for € = 1,2, … , ç

OP> … OPA gPP gP> … gP

- Consider the following matrices
OPP
O>P O>> … OPA g>P g>> … g>
­=ì ⋮ ⋮ ⋱ ⋮ ð and æ = ì ⋮ ⋱ ð
⋮ ⋮
OïP Oï> … OïA gAP gA> … gA
Rows of ­ are:
nP = £OPP OP> … OPA ¤
n> = £OPP OP> … OPA ¤
⋮
nï = £OïP Oï> … OïA ¤
Columns of æ are:
—P = £gPP g>P … gAP ¤
—> = £gP> g>> … gA> ¤
⋮
— = £gP g> … gA ¤

nP . —P nP . —> … nP . —
Then

n> . —P n> . —> … n> . —

­æ = ì ⋮ ⋮ ⋱ ⋮ ð = £—œ ¤ï×
nï . —P nï . —> … nï . —
Example: Find the product of the matrices

12 27 30 13
Solution:

AB = é ê
8 −4 26 12
Exercise: Determine the size of the product matrix ­æ if the sizes of ­ and æ are 4 × 5 and
5 × 7 respectively.

32
Note: The product of lower triangular matrices is lower triangular, and the product of upper
triangular matrices is upper triangular.

1) ­æ ≠ æ­, (matrix product is not commutative.)

Properties of Matrix Multiplication

2) ­(æ ) = (­æ) , (matrix multiplication is associative)

3) ­(æ + ) = (­æ + ­ ) Ode (æ + )­ = æ­ + ­, (multiplication of matrices is

4) If ­æ = T, it does not mean that either A = 0 or B = 0.

distributive with respect to addition )

10 00
Example: For matrix A and B given by­ = é ê æ = é ê we have
10 11
00
­æ = é êis a null matrix even though ­andæare not a null matrix.
00
5) The relation ­æ = ­ mn æ­ = ­ does not imply that æ = . In other words the

1 23 123 23 4
cancelation law doesn’t hold as for real numbers.

Example: if ­ = õ 1 1 2ö, æ = õ 1 1−1öand = õ2 2 0 ö

−14 3 22 2 111
9 10 7
We have, ­æ = õ6 7 6 ö = ­ , butæ ≠
9 8 −1
12 2 1 10
Exercise: If ­ = é êæ = é ê and = é ê, then verify that:
3 4 −3 2 21
a) A(BC) = (AB)C b) A(B + C) = AB + AC

Definition: If A is an ç × d matrix, then the transpose of A, denoted by ô , is defined to be the

Transpose of a matrix

matrix d × ç that results from interchanging the rows and columns of A.

Remark: The transpose of a row matrix is column matrix and the transpose of a column matrix
is a row matrix.
Example: The following are some examples of matrices and their transposes.

33
 1. (­ + æ)  = ­ + æ 
Properties of Matrix Transpose

2. (­ )  = ­
3. (­æ) = æ  ­ 

Proof: Ex.
Note: The transpose of a lower triangular matrix is upper triangular and the transpose of an

Orthogonal Matrix: An orthogonal matrix A is a matrix such that ­­ = ­ ­= I. A typical

upper triangular matrix is lower triangular.

1 1
orthogonal matrix is:

−
­ = Q2 Q2
1 1 
Q2 Q2

Definition: If A is a square matrix, then the trace of A, denoted by ‚n(­), is defined to be the
Trace of Matrix

sum of the entries on the diagonal of A. The trace of A is undefined if A is not a square matrix.
Example: The following are examples of matrices and their traces.

For any d × dsquare matrix ­ and for any polynomial,

Polynomial of Matrix

(`) = OA ` A + OAYP ` AYP + ⋯ + O where Oœ are scalars, wedefine (­) = OA ­A +

OAYP ­AYP + ⋯ + O þ.If (­) = (0œò ), then A is a zero (root) of the polynomial.
3−4
Examples: Let­ = é ê and let (`) = 2` > + ` + 3. Then compute (­).
11

34
Exercise:

1. Let ­ = é ê .Then compute:

10
a) (­) € (`) = 3` @ − 4` > + 2` b) (−3­) € (`) = 3` > − 2`

34
 Definition: Let ­ be a square matrix. Then ­is said to be
Symmetric and Skew-Symmetric Matrices:

Symmetric if ­ = ­ ,i.e. Oœò = Oòœ .

Skew-symmetric (anti-symmetric) if ­ = −­, i.e.Oœò = −Oòœ .
2 −1 3 0 5 −2 0 0 1
Example: A = õ−1 5 1öB = õ−5 0 3ö C = õ−5 0 3ö
3 1 7 2 −3 0 2 −3 0

Symmetric Skew-Symmetric Neither

3 g 0 3 −4
Exercise: Determine whether the following matrices are symmetric, skew symmetric or neither.

a) ­ = é ê
g 3 d) ý = õ−3 0 5 ö
1 0 0 4 −5 0
b) æ = é ê
0 0 1 0 0
e)  = é ê
0 5 −4 3 0 0

c) æ = ì−5 0 −7 −2ð
4 7 0 1
−3 2 1 0
2.5. Elementary Row Operations and Echelon Form of Matrices

A matrix ô is said to be row equivalent to matrix
, written ô~
if matrix
is obtained from ô by a
i. Elementary Row Operations

finite sequence of elementary row operations. These elementary row operations are:
i. Interchanging the € ° row by the ˜ ° row (i.e =œ ↔ =ò )
ii. Multiplying the € ° row by a none zero scalar (i.e=œ → ™=œ ).
iii. Replacing the € ° row by ™ times the ˜ ° row plus € ° row (i.e=œ → ™=ò + =œ )

2 6 1 5 7 −4 2 6 1 2 6 1
Example: Apply all elementary row operations on the given matrix:

ô = õ1 2 −1ö =P ↔ =@ õ1 2 −1ö => → 2=> õ2 4 −2ö => → 2=@ + => õ11 16 −9ö =

5 7 −4 2 6 1 5 7 −4 5 7 −4
3 −1 2
Exercise: Let ­ = õ 4 3 2ö. Then find matrix
which is row equivalent to ô with GUU =
−1 −2 7
T andG/q = T.
Remark: The first non-zero entry in a row is called the leading entry (pivotal entry) of that row.

35
 2 0 1
0 −2 −1
ì ð
−4 0 −3
0 0 0
Entries 2, −2 Ode − 4 are leading entries of nm˄ 1, 2,3 respectively, and no leading entry for row 3.

ii. Echelon Form of a Matrix

Definition: An m × n matrix ­is said to be in echelon form (EF) provided the following two conditions
hold.
1. Any zero rows (if there is) are at the bottom of the non-zero rows.
2. The leading entry of all rows is at the right side of the leading entries of the above rows (i.e. all

−1 2 0
entries below the leading entry are zero).

2 6 1 1 2 6 9 1 2 6 9
0 6 1
Example: ô = ì ð
= õ0 2 −2 1 ö  = õ0 1 7 6 ö  = õ0 1 7ö
0 0 −1
0 0 3 −4 0 0 0 −4 0 2 8
0 0 0
Matrices , and  are in echelon form, but matrix  is not.
Definition: An m × n matrix ­is said to be in row-echelon form (REF) provided the following two
conditions hold.
1. The matrix is in echelon form.

1 2 0 1 0 02
2. All leading entries are equal to 1.

1 6 1 1 1 6 9
0 1 1 0 1 0−5  = õ0 1 7ö
Example: ô = ì ð
= õ0 1 −2 1 ö = 
0 0 1 0 0 10
0 0 1 −4 0 1 8
0 0 0 0 0 00
Matrices ô,
Ode  are in row echelon form (REF) but  is not.
Definition3: An m × n matrix ­is said to be in reduced row-echelon form (RREF) provided the
following two conditions hold.
1. The matrix is in row echelon form.

1 0 0 1 0 −3 0 1 0 −30
2. All non-leading entries in a column, which contains the leading entries, are equal to 0.

1 0 0 0
0 1 0 0 1 0 0   = 0 1 5 0
Example: ô = ì ð
= õ0 1 0 0ö  = 
0 0 1 0 0 0 1 0 0 01
0 0 1 0
0 0 0 0 0 0 0 0 0 00
Remark: Any matrix can be reduced to its echelon form by applying some elementary row operations
on the given matrix.

36
Example: Reduce matrix ô to its row echelon form by applying elementary row operations where
3 −10 5
A = õ−1 12 2ö
1 −5 2
1 −5 2
Solution: Applying =P ↔ =@ we have: õ−1 12 2ö
3 −10 5
1 −5 2
Applying => → =P + => and =@ → =@ − 3=P we get:õ0 7 4ö
0 5 −1
1 −5 2
Applying => → we get:0 1 47
Õ
Z
0 5 −1
1 −5 2
47 ð
Applying =@ → =@ − 5=> we get:ì0 1
0 0 −277

1 −5 2
7
Appling =@ → − 27 =@ we get:0 1 47
0 0 1
1 −5 2
Applying => → => − we get:õ0 1 0ö
p
Z
0 0 1
This is in row echelon form.

Suppose an ë × H matrix is reduced by row operations to an echelon form !. Then the rank of ô,
Rank of a Matrix

denoted byèGH¦(ô), is defined to be,

Rank(ô) = number of pivots (leading entries) or
= number of nonzero rows in ! or
= number of basic columns in ô,
where the basic columns of ôare defined to be those columns in ôwhich contain the pivotal positions.
1 00 0 2
0 1 6 23
Example: Let matrix ô = ì ð. Then therank (A)=3 since matrix ô is in echelon form and has
0 01 00
0 0 0 00
three non-zero rows.

37
 1 2 1 1
Example: Determine the rank, and identify the basic columns in the matrix

­ = õ2 4 2 2ö
3 6 3 4
1 2 1 1 1 2 1 1 1 2 1 1
Solution:­ = õ2 4 2 2ö → õ0 0 0 0 ö → õ0 0 0 1ö
3 6 3 4 0 0 0 1 0 0 0 0
nOd™(­) = q and
U U
Basic Columns = ¸"q# , "q#$
/ †
Exercise: Reduce each of the following matrices to its echelon form and determine its rank & identify

1 2 3 3 1 2 3
the basic columns.

a. ­ = õ2 4 6 9ö 2 6 8

2 6 7 6 b. æ = 2 6 0
1 2 5
3 8 6
2.6. Inverse of a Matrix and Its Properties
Definition: If A is an d × d square matrix, and if a matrix B of the same size can be found such
that ­æ = æ­ = þ, where þ the d × didentity matrix, then A is said to be invertible (non-singular)and
B is called an inverse of A. If no such matrix B can be found, then A is said to be singular.
Example: The matrix

Since

And

Remark:
1. ­YP ≠ %
P

2. Matrices which are not square matrices have no inverses

3. Not all square matrices have inverse
4. (­YP ) = (­ )YP

38
Properties of Inverse Matrices
1. If B and C are both inverses of a matrix A, then B = C. that is the inverse of a matrix is unique
and is denoted by AYP .
2. If A and B are invertible matrices of the same size, then
a) AB is invertible and
b) (­æ)YP = æ YP ­YP
3. ­YP is invertible and (­YP )YP = ­
eP 0 … 0
0 e> … 0
4. Let ý = ì ð be an d × d diagonal matrix where all eœ ≠ 0 (for € = 1, 2, … , d).
⋮ ⋮ ⋱ 0
0 0 … eA
0 … 0
P
П

0 … 0
P
Then the inverse of ý, denoted by ý , is given by ý
YP YP
= ÐÕ
⋮ ⋮ ⋱ 0

0 0 …
P
Ð&

Exercise: Verify that ýý YP = ýYP ý = þ.

Finding Inverse of a Matrix by Using Elementary Row Operations (Gauss-Jordan Elimination

Let ôbe an d × dmatrix and let 'H be an d × d identity matrix. Then to find the inverse of ô.
Method)

1. Adjoin the identity d × dmatrix 'H to ôto form the augmented matrix (ô: 'H ).
2. Compute the reduced echelon form of (ô: 'H ). If the reduced echelon form is of the type ('H :
),
then æ is the inverse of ô. If the reduced echelon form is not of the type ('H :
), in that the first
d × dsub matrix is not 'H , then ôhas no inverse.

1 2 3
Example: Find the inverse of A if

A = õ2 5 3ö
1 0 8
1 2 3⋮1 0 0
Solution:£­: þ@ ¤ = õ2 5 3⋮0 1 0ö
1 0 8⋮0 0 1
Applying => → => − 2=P we get:

39
 1 2 3⋮1 0 0
õ0 1 −3⋮−2 1 0ö
1 0 8⋮0 0 1
Applying =P → =P − 2=> we get:
1 0 9 ⋮ 5 −2 0
õ0 1 −3⋮−2 1 0ö
1 0 8⋮0 0 1
Applying =@ → =@ − =P we get:
1 0 9 ⋮ 5 −2 0
õ0 1 −3⋮−2 1 0ö
0 0 −1⋮−5 2 1
Applying =P → =P + 9=@ ; => → => − 3=@ and =@ → −=@ we get:
1 0 0⋮−40 16 9
õ0 1 0⋮−13 −5 −3ö
0 0 1⋮ 5 −2 −1
−40 16 9
Therefore, B = õ−13 −5 −3ö = ­YP
5 −2 −1

1 0 2
Example: Find the inverse of ­ = (2 −1 3) by using elementary row operations (Gauss-Jordan
4 1 8
Method).

1 0 2⋮1 0 0
£­ ⋮ þ@ ¤ = õ2 −1 3⋮0 1 0ö
4 1 8⋮0 0 1
Solution:

Applying=> → => − 2=P ,we get:

1 0 2⋮1 0 0
õ0 −1 −1⋮−2 1 0ö
4 1 8⋮0 0 1
Applying=@ → =@ − 4=P we get:
1 0 2⋮1 0 0
õ0 −1 −1⋮−2 1 0ö
0 1 0 ⋮−4 0 1
Applying =@ → => +=@ we get:
1 0 2⋮1 0 0
õ0 −1 −1⋮−2 1 0ö
0 0 −1⋮−6 1 1
Applying=@ → −=@ we get:

40
 1 0 2⋮1 0 0
õ0 −1 −1⋮−2 1 0ö
0 0 1 ⋮ 6 −1 −1
Applying=P → =P − 2=@ and => → => + =@ we get:
1 0 0⋮−11 2 2
õ0 −1 0⋮ 4 0 −1ö
0 0 1⋮ 6 −1 −1
Applying => → −=> we get:
1 0 0⋮−11 2 2
õ0 1 0⋮ −4 0 1ö
0 0 1⋮ 6 −1 −1
1 0 0⋮−11 2 2
£þ@ ⋮ æ¤ = õ0 1 0⋮ −4 0 1ö
0 0 1⋮ 6 −1 −1
−11 2 2
⇒ æ = õ −4 0 1 ö = ­YP
6 −1−1

1 −1 −2
Example: Determine the inverse of the matrix

ô=õ 2 −3 −5ö
−1 3 5
1 −1 −2 ⋮ 1 0 0
(­: '/ ) = õ 2 −3 −5 ⋮ 0 1 0ö
−1 3 5 ⋮ 0 0 1
Solution:

Appling => → => −2=P & =@ → =@ + =P we get

1 −1 −2 ⋮ 1 0 0
õ0 −1 −1 ⋮ −2 1 0ö
0 2 3 ⋮ 1 0 1
Appling => → −=> we get
1 −1 −2 ⋮ 1 0 0
õ0 1 1 ⋮ 2 −1 0ö
0 2 3 ⋮ 1 0 1
Appling =P → => +=P & =@ → =@ − 2=> we get
1 0 −1 ⋮ 3 −1 0
õ0 1 1 ⋮ 2 −1 0 ö
0 0 1 ⋮ −3 2 1
Appling =P → =@ +=P & => → => − =@ we get
1 0 0 ⋮ 0 1 1 0 1 1
õ0 1 0 ⋮ 5 −3 −1ö Thus ôYU = õ 5 −3 −1ö
0 0 1 ⋮ −3 2 1 −3 2 1

41
 1 1 5 1 2 −3 2 1 3 1 6 4
Exercise: Determine inverse of the following matrices, if it exists.

­ = õ1 2 7 ö
= õ1 2 1ö = õ0 2 1ö ý = õ 2 4 −1ö
2 −1 4 5 −2 −3 1 1 2 −1 2 5
2.7. Determinant of a Matrix and Its Properties

of a matrix variable in the sense that it associates a real number (­) with a square matrix A. Our work
Introduction: In this section, we shall study the “determinant function,” which is a real-valued function

on determinant functions will have important applications to the theory of systems of linear equations
and will also lead us to an explicit formula for the inverse of an invertible matrix.

matrixô, and it is usually denoted by I+º(ô) orDôD. i.e. if ô is an d × d square matrix, then
Definition: A “determinant” is a certain kind of function that associates a real number with a square

OPP OP> … OPA

O>P O>> … O>A
det(­) =  ⋮ ⋮ ⋱ ⋮ 
OAP OA> … OAA
a. Determinant of U × U , Gºè“t+v
Let ­ = £OPP ¤be 1 × 1 matrix. Then det(­) = OPP.
Example: Let ­ = £2¤. Then det(­) = 2
b. Determinant of q × q ,Gºè“t+v
OPP OP> OPP OP>
Let ­ = éO ê × 2 det(­) = ¢ O>> ¢ = OPP O>> − OP> O>P
>P O>> O>P
be a2 matrix. Then

1 2
Example: Find det(­) if ­ = é ê
−2 0
Solution: det(­) = 1(0) − 2(−2) = 4
c. Determinant of H × H matrix
Definition-1: Let ô be an d × dsquare matrix and ,“¥ be the (d − 1) × (d − 1) matrix obtained from
matrix ô by deleting € ° row and ˜ ° column containing the entry Oœò . Then -.Î (,“¥ ) is called the
minor of OϘ .
Remark: the matrix ,“¥ is called sub matrix of ô.
Definition-2: The cofactor of Oœò , denoted by œò , is defined as œò = (−1)œÄò det (/œò ).
2 5 −4
Example: Let ­ = õ3 −1 2 ö. Then find the minor and cofactor of OPP and O>@ .
5 4 6

42
 −1 2
Solution: Minor of OPP = det(/PP ) = ¢ ¢ = −1(6) − 2(4) = −14
4 6
Cofactor of OPP = PP = (−1)PÄP det(/PP ) = −14
2 5
Minor of O>@ = det(/>@ ) = ¢ ¢ = 2(4) − 5(5) = −17
5 4
Cofactor of O>@ = >@ = (−1)>Ä@ det(/>@ ) = 17
Definition-3: The determinant of an d × d matrix ­ is given by either of the following two formulas.
i. det(­) = OœP œP + Oœ> œ> + ⋯ + OœA œA = AòP Oœò œò , for fixed € = 1,2,3, …
ii. det(­) = OPò Pò + O>ò >ò + ⋯ + OAò Aò = AœP Oœò œò , for fixed ˜ = 1,2,3, …

0 2 3 0
Example: Evaluate the determinant for the matrices

2 −1 3
0 4 5 0
­ = õ1 2 4ö and æ = ì ð
0 1 0 3
5 −3 6
2 0 1 3
Solution: Let us take row-1 of matrix ­ for €. Then
A @

-.Î(ô) = 0 Oœò œò = 0 OPò Pò = OPP PP + OP> P> + OP@ P@ = 2(24) − 1(14) + 3(−13) = −‘
òP òP

Let us take column-1 of matrix æ for ˜.

A p

-.Î(
) = 0 Oœò œò = 0 OœP œP = OPP PP + O>P >P + O@P @P + OpP pP

œP œP

=0 PP +0 >P +0 @P + 2(6) = Uq
Example: Let ­ = (OP , O> , O@ ), æ = (gP , g> , g@ ) Ode = (—P , —> , —@ ) be vectors in = @ . Then show
OP gP —P
that¡O> g> —> ¡ = ­ ⋅ (æ × )
O@ g@ —@
OP gP —P
g g@ g g@ g g>
Solution: ¡O> g> —> ¡ = OP  >  − O>  P  + O@  P 
—> —@ —P —@ —P —>
O@ g@ —@
= OP (g> —@ − g@ —> ) − O> (gP —@ − g@ —P ) + O@ (gP —> − g> —P )
= (OP , O> , O@ ) ⋅ (g> —@ − g@ —> , gP —@ − g@ —P , gP —> − g> —P )
= ­ ⋅ (æ × )
OPP OP> OP@
Note: The determinant of a 3×3 square matrix ­ = õ >P O>> O>@ öis given by
O
O@P O@> O@@

43
 OPP OP> OP@
O>>O>@ O>P O>@ O>P O>>
e‚(­) = ¡O>P O>> O>@ ¡ = OPP ¢O O ¢ − OP> ¢O O ¢ + OP@ ¢O O ¢
O@P O@> O@@ @> @@ @P @@ @P @>

Letô,
and C be H × H square matrices and ¦ be any scalars. Then

Properties of Determinants

a. det(­) = det (­ )

b. det(™­) = ™ A det (­), where d is the size of ­.
c. det(­ + æ) ≠ det(­) + det (æ)
d. det(­æ) = det(­) det (æ)

e. det(­YP ) = if ­ is invertible (non-singular) matrix.

P
123 (%)
,

f. Determinant of any d × d triangular matrix ­ is the product of its diagonal entries.

i.e. det(­) = OPP O>> … OAA
g. Determinant of d × d diagonal matrix ýis the product of its diagonal entries. Thus det(þA ) = 1,
where þA is an d × d identity matrix.
h. If any rows (or columns) of a square matrix ­ are proportional to each other (i.e. one is the scalar
multiple of the other), then det(­) = 0.
i. If any row (or column) of a square matrix ­ is zero, then det(­) = 0.
Exercise:

2 16 40 16 14 1 2 −3 4 5
1. Evaluate the determinant for each of the following matrices.

0 −3 22 −3 −18 Q2 0 0 0 4 −5 6 5 −2
 
­ = 0 0 4 1 20 , æ = ì−8 Q2 0 0ð, = 7 8 −8 6 20 , ý=
0 0 0 5 0 7 0 −1 0 3 6 −9 12 15
0 0 0 0 1 9 5 0 1 0 −2 3 4 1
1 0 6 2 0 0
õ3 5 2ö,  = õ0 −1 0ö
0 0 0 0 0 2
2. Let det(­) = 4 and det(æ) = −3,and let ­ and æ be a 5 × 5 matrices. Then find
i. det (2­) iii. det (­æ)
det (­ ) det (­YP )
` − 1 −2
ii. iv.

3. Let ­ = é êand let det(­) = 0. Then find the value/s of `.

`−2 `−1

44
Theorem-1: Let A be an invertible matrix. Then e‚(­YP ) = (e‚(­))YP.
Proof: þ = ­­YP then taking determinant on both sides we get:
e‚(þ) = e‚(­­YP )
⇒ 1 = e‚(­)e‚(­YP )
1
⇒ e‚(­YP ) = = (e‚(­))YP
e‚(­)
2 1
Example: If ­ = é ê, then e‚(­YP ) = = @.
P P
3 3 Ð4(%)

Note: We can evaluate the determinant of any square matrix ­ by reducing it to its echelon form by
keeping the following conditions.
a. If æ is a matrix which results by applying the elementary row operation of multiplying any
particular row of ­ by a non-zero constant ™, then det(æ) = ™e‚(­).
b. If æ is a matrix that results when two rows or two columns of ­ are interchanged, thendet(æ) =
−det (­).
c. If æ is a matrix that results when a multiple of one row of ­ is added to another row or when a
multiple of one column is added to another column, then det(æ) = det (­).
O g —
Exercise: Let­ = õe   ö and let det(­) = −12. The find det (æ), det ( )and det (ý) if
¬ ℎ ™
3O 3g 3— O g — O g —

=õe   ö,  = õ¬ ℎ ™ö and ý = õe + 2O g + 2g  + 2— ö
¬ ℎ ™ e   ¬ ℎ ™

OPP OP> ⋯ OPA

Adjoint and Inverse of a Matrix

O>P O>> ⋯ O>A

Definition-1: Let ­ = ì ⋮ ⋮ ⋱ ⋮ ðbe and × dsquare matrix and let —œò be the cofactor of
OAP OA> ⋯ OAA
—PP —P> ⋯ —PA
—>P —>> ⋯ —>A
Oœò .Then the matrix =ì ⋮ ⋮ ⋱ ⋮ ðwhose (€˜) entry is —œò is called matrix of cofactors of
°

—AP —A> ⋯ —AA

—PP —>P ⋯ —AP
—P> —>> ⋯ —A>
entries of ­ and its transpose 
=ì ⋮ ⋮ ⋱ ⋮ ðis called the adjoint of ô, denoted by GI¥(ô).
—PA —>A ⋯ —AA

45
 34 6
Example: Find the matrix of cofactors and adjoint of the matrix

­ = õ3 9 0 ö
1 2−1
Solution:

90 34
The cofactors of each elements of A are

= (−1)PÄP det (/PP ) = (−1)PÄP ¢ ¢ = −9 = (−1)>Ä@ det (/>@ ) = (−1)>Ä@ ¢ ¢ = −2

PP
2−1 >@
12
30 46
P> = (−1) e‚(/P> ) = (−1)PÄ> ¢ ¢=3 = (−1)@ÄP det (/@P ) = (−1)@ÄP ¢ ¢ = −54
PÄ>
1−1 @P
90
39 36
P@ = (−1) det (/P@ ) = (−1)PÄ@ ¢ ¢ = −3 = (−1)@Ä> det (/@> ) = (−1)@Ä> ¢ ¢ = 18
PÄ@
12 @>
30
46 34
>P = (−1) det (/>P ) = (−1)>ÄP ¢ ¢ = 12 = (−1)@Ä@ det (/@@ ) = (−1)@Ä@ ¢ ¢ = 15
>ÄP
20 @@
39
36
>> = (−1) det (/>> ) = (−1)>Ä> ¢ ¢ = −9
>Ä>
1−1
PP P> P@ −9 3 −3
=õ >P >> >@ ö = õ 12 −9−2ö
−54 18 15
Matrix of Cofactors=
@P @> @@

−9 12 −54
-5() = 
= õ 3 −9 18 ö
−3−2 15

3 2 −1
Exercise: Find the adjoint of A if:

­ = õ1 6 8ö
2 −4 0
Definition-2: Let ­ be and × d square matrix with det (­) ≠ 0. Then the inverse of a matrix ­, denoted

by ­YP , is defined as ôYU = -.Î(ô) GI¥(ô).

U

Note: If det(­) = 0, then the matrix ­ has no inverse (i.e. ­ is singular matrix).
34 6
Example: Find the inverse of the matrix ­ = õ39 0 ö
12−1
Solution: det(­) = −33
−9 12 −54
Adj(A) = õ 3 −9 18 ö
−3−2 15

46
 3 −4 19

 11 11 11
1 1 −9 12 −54 1 3 −6
­YP = Oe˜(­) = − õ 3 −9 18 ö = −
det(­) 33  11 11 11
−3 −2 15
 1 2 −5
 11 33 11

1 2 3 2 −1 3
Exercise: Verify whether each of the following matrices has an inverse, and find the inverse if it has.

1 4
­=é ê, æ = õ 0 −1 2ö and ý = õ1 2 4ö
3 2
−4 5 0 2 4 8

2.8. System of Linear Equations

A system of ç linear equations with n-unknown (variables) is given by
OPP `P + OP> `> + ⋯ + OPA `A = gP
O>P `P + O>> `> + ⋯ + O>A `A = g>
⋮ ⋮ ⋱ ⋮ ⋮
OïP `P + Oï> `> + ⋯ + OïA `A = gï

OPP OP> … OPA `P gP

The matrix form of this system of linear equations is

O>P O>> … O>A `> g>

ì ⋮ ⋮ ⋱ ⋮ ðì ⋮ ð = ì ⋮ ð
OïP Oï> . . OïA `A gï

ô¹ = ¼
A x b

Where, ­ is the coefficient matrix, ` is the unknown matrix (vector) and g is the known matrix (vector).
The augmented matrix, £­ ⋮ g¤, for the above system of linear equation is
OPP OP> … OPA ⋮ gP
O>P O>> … O>A ⋮ g>
ì ⋮ ⋮ ⋱ ⋮ ⋮ ⋮ ð
OïP Oï> . . OïA ⋮gï
Remark: If a system of linear equations is the form ô¹ = T, i.e. all entries of g are equal to 0, then the
system is called homogeneous system of linear equation, otherwise it is called non-homogeneous.
Example: write the coefficient matrix and the augmented matrix for the following system of linear
equations.

47
 3`P − `> + `@ = 4
`P + `> + `@ = 6
`P − `> − `@ = −4
3 −1 1
Solution: Coefficient Matrix = ­ = õ1 1 1ö
1 −1 −1
3 −1 1 ⋮ 4
Augmented Matrix= £­ ⋮ g¤ = õ1 1 1 ⋮ 6ö
1 −1 −1 ⋮ −4
2.9. Solving System of Linear Equations

Let ô¹ = ¼ be a system of d linear equations in d unknowns such that det(­) ≠ 0. Then the system has
I. Cramer’s Rules

a unique solution. This solution is:

`P = , `> = , …, `A =
123 (%Ÿ ) 123 (%Õ ) 123 (%& )
123 (%) 123 (%) 123 (%)

Where ­œ (for € = 1, 2, … , d) is the matrix obtained by replacing the entries in the € ° column of ­ by the
gP
g>
entries in the matrix g = 6 7.
⋮
gA

`P + `@ = 6
Example: Using Cramer’s Rule, solve the following system of linear equations.

−3`P + `> + `@ = 30
−`P − `> + `@ = 8

1 0 2 1 6 2
Solution:

­ = õ−3 4 6ö ­> = õ−3 30 6ö

−1 −2 3 −1 8 3
6 0 2 1 0 6
­P = õ30 4 6ö ­@ = õ−3 4 30ö
8 −2 3 −1 −2 8
det(­) = 44, det(­P ) = −40, det(­> ) = 72, and det(­@ ) = 152

Thus ¹U = = = ¹q = = pp = UU and ¹/ = = = UU
123 (%Ÿ ) Yp· YUT 123 (%Õ ) Z> U‹ 123 (% ) PX> /‹
123 (%) pp UU 123 (%) 123 (%) pp

48
 5` + 7a = 12 2`P − `> − `@ − `p = 6
Exercise: Solve the following system of linear equations using Cramer’s Rule:

a) ¸10` + a + 3b = 14› `P − 5`> − 3`@ − `p = 1

` + 6a + 2b = 9 5`P + `> − 7`@ + 6`p = −3
b)
−`P − `> − `@ = 3
II. Gaussian Elimination Method
Definition: The process of using elementary row operations to transfer an augmented matrix of linear

Let­` = g be a system of linear equations. Then, to solve the system by using Gaussian elimination
system in to one whose augmented matrix is in row echelon form is called Gaussian elimination.

method, use the following procedures.

I. Write down the augmented matrix for the system.
II. Reduce this augmented matrix to its row echelon form.
III. Use back substitution to arrive at the solution.

2`P − `> − `@ − `p = 6
Example: By using the Gaussian-Elimination method, solve the following system of linear equations.

`P − 5`> − 3`@ − `p = 1
5`P + `> − 7`@ + 6`p = −3
−`P − `> − `@ = 3

4`P + `> + `@ + `p = 6 −`P − 4`> + 2`@ + `p = −32

Exercise:By using the Gaussian-Elimination method, solve the following system of linear equations.

3`P + 7`> − `@ + `p = 1 2`P − `> + 7`@ + 9`p = 14

7`P + 3`> − 5`@ + 8`p = −3 −`P + `> + 3`@ + `p = 4
`P − 2`> + `@ − 4`p = −4
i. ii.
`P + `> + `@ + 2`p = 3

Theorem: If ô is an invertible d × d matrix, then for each d × 1 matrix
, the system ­` = g has
III. Inverse Matrix Method

exactly one solution namely ` = ­YP g = g

%Ðò%
Ð4%

`P + 2`> + 3`@ = 5
Example: Solve the following system of linear equations by using inverse method.

¸ P + 5`> + 3`@ = 3 ›
2`
`P + 8`@ = 17
In matrix form, this system can be written as:­` = g where
1 2 3 `P 5
A = õ2 5 3ö ` = õ`> ö and g = õ 3 ö
1 0 8 `@ 17

49
 −40 16 9
But ­YP = õ−13 −5 −3ö and we have:
5 −2 −1

−40 16 9 5 1 `1
` = ­ g = õ−13
YP
−5 −3ö õ 3 ö = õ−1ö = õ`2 ö
5 −2 −1 17 2 `3
Thus`P = 1, `> = −1 Ode `@ = 2 On ‚ℎ „mÀ’‚€md„

`+a=2 `P + 3`> + `@ = 4
Exercise: Solve the following system of linear equations by using inverse method.

5` + 6a = 9 b) 2`P + 2`> + `@ = −1
a)
2`P + 3`> + `@ = 3
Remark:-
1. A system of equations that has no solution is called inconsistent; if there is at least one solution
of the system, it is called consistent.
2. A system of linear equations may not have solutions, or has exactly one solution (unique

3. Every homogeneous system of linear equations (€. , . ô¹ = T) is consistent, since all`œ = 0 (for
solution), or infinitely many solutions.

€ = 1, 2, 3, … ) is a solution. This solution is called the trivial solution; if there are other solutions,
they are called nontrivial solutions. If this system has nontrivial solutions, then these solutions are
infinite.
4. If the number of variables is greater than the number of equations in a given system of linear
equations, then the system has infinite solutions. The arbitrary values that are assigned to the free

5. Let ­` = g be a system of non-homogenous linear equations and the number of variables are
variables are often called parameters.

equal to the number of equation (i.e. ­ be a square matrix). And let ­œ (for € = 1, 2, … , d) be a
matrix obtained by replacing the entries in the € ° column of ­ by the entries in the matrix
gP
g>
g = 6 7. Then
⋮
gA
i. if det(­) ≠ 0, then the system has a unique solution.
if det(­) = 0, then the system has
a. infinitely many solutions if det(­œ ) = 0 for all € = 1,2,3 …
ii.

50
 b. no solution if at least one of the det(­œ ) ≠ 0 for some € = 1,2,3 …
2.10. Eigen values and Eigenvectors
Definition: Let ­ be an d × d square matrix and ` be a non-zero column vector. Then `is called the
eigenvector ( or right eigenvector or right characteristic vector) of ­ if there exists a scalar 8 such that
­` = 8` … . . (1)
Then, 8 is called an eigenvalue or characteristics value of ­. Eigenvalue may be zero, but eigenvector

1
cannot be zero vectors.

Example: show that é ê is an eigenvector corresponding to the eigenvalue 8 = −2 for the matrix
−1
3 5
é ê.
−2 −4

3 5 1 2 1
Solution: from equation (1) we have

é ê é ê = é ê = −2 é ê
−2 −4 −1 −2 −1
From equation (1), ­` − 8` = (­ − 8þA )` = 0

D­ − 8þA D is called the characteristic polynomial of ­.

The equation D­ − 8þA D = 0 is called the characteristic equation.

For each eigenvalue 8, the corresponding eigenvector is found by substituting 8 back into the
characteristic equation D­ − 8þA D = 0.
3 5
Example: Determine the eigenvalues and corresponding eigenvectors of the matrix ­ = é ê
−2 −4
3 5 1 0 3−λ 5
Solution: For this matrix ­ − λI = é ê − λé ê=é ê and hence
−2 −4 0 1 −2 −4 − λ
det(­ − λI) = (3 − λ)(−4 − λ) − 5(−2) = λ> + λ − 2 = 0. Thus, the characteristic equation of A is
λ> + λ − 2 = 0 and up on solving this we get λ = 1 and λ = −2
`P
The eigenvectors to : = U will be obtained by solving equation (1) above for ¾ = é` ê. With
>
i.

this value of λ after substituting and rearranging, we get:

3 5 1 0 `P 0 3 5 `P 0
ªé ê− 1é ê« é ê = é ê ⇒é êé ê = é ê
−2 −4 0 1 `> 0 −2 −4 `> 0

2`P + 5`> = 0
This is equivalent to the set of linear equations given below:

−2`P − 5`> = 0

51
The solution to this system is `P = `> with `> arbitrary, so the eigenvectors corresponding to λ = 1
YX
>

are,
`P YX
`
YX
¾ = é` ê = ; > < = `> ; > < with `> arbitrary.
>
> `> 1
When : = −q, equation (1) above may be written as:
3 5 1 0 `P 0 5 5 `P 0
ii.

ªé ê − (−2) é ê« é` ê = é ê ⇒é ê é` ê = é ê
−2 −4 0 1 > 0 −2 −2 > 0

5`P + 5`> = 0
This is equivalent to the set of linear equations given below:

−2`P − 2`> = 0
The solution to this system is `P = −`> with `> arbitrary, so the eigenvectors corresponding to λ = −2

`P −`> −1
are,

¾ = é` ê = é ` ê = `> é ê with `> arbitrary

> > 1

5 2 2
Exercise: Determine the eigenvalue and eigenvector of the following matrices if there exist:

­ = õ3 6 3ö
6 6 9
3 −2 0
æ = õ−2 3 0ö
0 6 5

52