ENM 205 LINEAR ALGEBRA

Textbook: Kolman B., Hill D.R., (2000),

Elementary Linear Algebra, 7th Ed., Prentice
Hall

• COURSE PLAN
• 1.INTRODUCTION, LINEAR EQUALITIES: Basic concepts, system of linear
equations (Slides 1-18=18)
• 2. MATRICES: Matrices, matrix operations and their algebraic properties.(18-
51=33))
• 3. MATRICES: Special types of matrices and partitioned matrices.(52-71=19)
• 4. MATRICES: Echelon form of matrices and elementary matrices.(72-108=36)
• 5. MATRICES: Inverse of matrices, equivalent matrices, LU factorization.(108-
127=19)
• 6. REAL VECTOR SPACES: Vectors in plane and space, vector spaces.
• 7. MIDTERM I
• 8. REAL VECTOR SPACES: Span and linear independency.
• 9. REAL VECTOR SPACES: Basis and dimension.
• 10 Rn VECTOR SPACE: Homogen systems
10. systems, coordinates and isomorphism
isomorphism.
• 11. Rn VECTOR SPACE: Rank of a matrix.
• 12. MIDTERM II
• 13. DETERMINANTS: Definition, properties of determinants, inverse of a matrix.
• 14. DETERMINANTS: Other applications of determinants, determinants with
regard to computations

1
 1. LINEAR EQUATIONS AND
MATRICES
• 1.1 Systems of Linear Equations

Linear Equation:

A linear equation in n unknowns, x1, x2,....,xn, is an equation of the form

“a1x1+a2x2+.....+anxn=b” ......(1)

where a1, a2,....,an, b are given real numbers.

xj, j=1,2,...n
j 12 are refered
f d tot variables
i bl or the
th unknowns.
k
aj, j=1,2,...,n are the constants.
b is also a constant.

In many problems we are given b and must find numbers x1, x2,.....,xn
satisfiying (1)

Solution to Linear Equation (1)

is a sequence of n numbers s1, s2,...,sn which has the

property that (1) is satisfied when x1=ss1,
x2=s2,.....xn=sn are substituted in (1).

Example:

The linear equation 6x1-3x

3x2+4x3=-13
13,

x1=2, x2=3 and x3=-4 is a solution to the linear

equation since 6(2)-3(3)+4(-4)=-13

2
• Linear System (Linear Equations Sytem):
A linear system is a set of m linear equations each in n unknowns.

a11x1 + a12x2 + .... + a1nxn = b1

a21x1 + a22x2 + .... + a2nxn = b2
# # + .... + # = # (2)
ai1x1 + ai2x2 + .... + ainxn = bi
# # + .... + # = #
am1x1 + am2x2 +....+ amnxn = bm

The closed form of the expression (2)

n
∑ a ij x j = bi where
h i=1,2,...,m
i 12
j =1

• aij, i=1,2,...,m and j=1,2,...,n are known constants

• Given values of bi we want to find values of x1, x2,..., xn that will satisfy each equation
in (2).

• A solution to linear system (2) is a sequence of n numbers s1, s2,..., sn which

has the property that each equation in (2) is satisfied when x1=s1, x2=s2,...,
xn=sn are substituted.

y
• If the linear system (2)
( ) has no solution, it is said to be INCONSISTENT.

• If it has a solution, then it is called CONSISTENT.

• If b1=b2=.....=bm=0 then (2) is called a HOMOGENEOUS SYSTEM.

• x1=xx2=.....=xxn=00 is always a solution to a homogeneous system,

system it is called
the TRIVIAL SOLUTION.

• A solution to a homogeneous system in which not all of x1, x2,...,xn are zero
is called a NONTRIVIAL SOLUTION.

3
EQUIVALENT LINEAR SYSTEMS
•Consider another system of r linear equations in n unknowns (3):

c11x1 + c12x2 +....+ c1nxn = d1

c21x1 + c22x2 +....+ c2nxn = d2
# # +....+ # = # (3)
ci1x1 + ci2x2 +....+ cinxn = di
# # +....+ # = #
cr1x1 + cr2x2 +....+ crnxn = dr

We say that (2) and (3) are equivalent if they both have exactly the same
solutions.

Example (equivalent linear systems)

The linear system The linear system

x1 - 3x2 = -7 8x1 - 3x2=7
2x1+ x2 = 7 3x1 - 2x2 = 0
10x1 - 2x2 = 14
has the only solution has the only solution
x1=22 andd x2=33 x1=22 andd x2=33

Thus these two systems are equivalent.

4
• Example (Homogeneous Sytems)

2x1 - x2 + 2x3 + x4 - 2x5 = 0

x1 + x2 - x3 + x4 =0
x2 + 3x3 - x5 = 0

Trivial Solution to the system

{x1=0, x2=0, x3=0, x4=0, x5=0}

A Nontrivial Solution to the system

{x1=7/13, x2=-2/13, x3=5/13, x4=0, x5=1}

THE METHOD OF ELIMINATION

•Example : Find the values of x1 and x2 for the following linear system
x1 - 3x2 = -3
2x1 + x2 = 8

To eliminate x1, we add -2 times the first equation to the second one
-2/ x1-3x2 =-3
2x1+ x2 = 8
-2x1+6x2 =6
2x1+ x2 = 8
7x2=14 Thus we obtain this equation having no x1 term.
x2=2 (i.e. We eliminated the unkonwn x1)

Then solving for x2, we have x2=2 and substituting into the first equation of
the linear system. We obtain x1-3(2)=-3 then x1=3

5
• Example (Inconsistency):

Lets consider the following L.E.S(linear equation system)

x1 - 3x2 = -7
2x1 - 6x2 = 7

-2/ x1 - 3x2 =-7

2x1 - 6x2 = 7
-2x1 +6x2 =14
2 1 - 6x
2x 6 2=7
0 = 21 This conclusion has no sense, that means that
the considered LES has no solution; it is inconsistent

•Example (Equivalent equation systems)

Lets consider (4) and (5) to find
x2 and x3 values:
-7x2 - 4x3 = 2
x1+2x2+3x3= 6 (1) -x2 - 2x3 = -4
2x1-3x2+2x3=14 (2) -7x2 - 4x3 = 2
3x1+ x2- x3= -2 (3) -7/ -x2 - 2x3 = -4
-7x2 - 4x3 = 2
7x2 + 14x3 = 28
To eliminate x1 we add -2 times the 1st equation to 10x3=30
the second one and -3 times the first equation to the
third one. x3=3

Substituting the value of x3

into the (5),
-2/ x1+2x2+3x3= 6 -3/ x1+2x2+3x3= 6
-x2-2(3)=-4 then we
2x1-3x2+2x3=14 3x1+ x2- x3= -2 obtained x2=-2
-2x1- 4x2- 6x3=-12 -3x1-6x2- 9x3 = -18
2x1-3x2+2x3= 14 3x1+ x2- x3= -2 Then substituting these
-7x2- 4x3 = 2 (4) -5x2-10x3=-20 values of x2 and x3 into the
-x2- 2x3=-4 (5) (1),
x1+2(-2)+3(3)=6 then
we find out x1=1

6
NOTE THAT
x1+2x2+3x3= 6 (1)
-x2- 2x3= -4 (5)
x3=3
This LES is equivalent to the LES which is given
at the beginning of the example . Why?

• Example (Infinitely many

solutions) Substituting x2 into the first equation:
x1 +2x2 - 3x3 = -4 (1)
x1 +2x2 - 3x3 = -4 (1) x1+ 2(x3-4) -3x3 = -4
2x1 + x2 - 3x3 = 4 (2) x1+ 2x3-8 -3x3 =-4 we
obtain
x1-x3=4 ((4))
To eliminate x1, lets add -2 times
the first equation to the second We get the x1 value in terms of x3 as
equation. We obtain: x1=x3+4

-2x1 -4x2 +6x3 = -4

2x1 + x2 - 3x3 = 4 Thus, a solution to the LES is
-3x2+3x3=12 x1=x3+4
-x2+x3=4 (3) x2=x3-4
x3 is
i any reall number
b
This means that the LES has
We have to solve equation (3) now. INFINITELY MANY
SOLUTIONS
A solution to (3) is x2=x3-4 where
x3 can be any real number.

7
 ANALYTICAL INVESTIGATIONS
OF SOLUTION CASES
• Unique solution
• No solution
• Infinitely many solutions

Analytical investigation of
unique solution case
• Example x2
A UNIQUE SOLUTION
x1 - 3x2 = -3 (1) 8 ℓ2
2x1 + x2 = 8 (2)

Solution is a
-2/ x1-3x2 =-3 point
ℓ1
2x1+ x2 = 8
-2x1+6x2 = 6 1
2x1+ x2 = 8
x1
7x2=14 -3 4
x2=2
x1=3

The graph of each of these equations is a straight line which we denote by ℓ1 and ℓ2,
respectively. The solution of this LES occurs at the intersection point of these two lines.
1) x1 0 -3 2) x1 0 4
x2 1 0 x2 8 0

8
 Analytical investigation of no
solution case
x2
•Example NO SOLUTION
x1 - 3x2 = -7 ℓ1
2x1 - 6x2 = 7
2,3
ℓ2
-2/ x1- 3x2 = -7
2x1 -6x2 = 7
-7 3,5
-2x1+6x2 =14 -1,17
2x1-6x2 = 7
0=21 it is inconsistent
No solution

The ggraph
p of these equations
q are parallel
p lines which we denote byy ℓ1 and ℓ2 respectively.
p y There is
no intersection point.
1) x1 0.0 -7 2) x1 0.00 3.5
x2 2.3 0 x2 -1.17 0.0

Another example x1+ 2x2=3

2x1+4x2=4

Analytical investigation of
infinitely many solutions case
x2
Example INFINITELY MANY
SOLUTIONS

x1+2x2=4
2x1+4x2=8

x1= 4-2x2 2
x2 is any real mumber
x1
4

The graph of these equations are overlapping lines which we denote by ℓ1 and ℓ2
respectively.
i l OnO a line
li there
h are infinitely
i fi i l many points.
i So
S all
ll the
h points
i in i that
h
lines are solutions to the LES.

1) x1 0 4 2) x1 0 4
x2 2 0 x2 2 0

9
 SUMMARY
• The elimination method has been described in general terms. We did not
indicate any rules for selecting the unknowns to be eliminated.
• Before pproviding
g a veryy systematic
y description
p of the method of elimination,,
we introduce, in the next section, the notation of a matrix.
• Matrix notation will greatly simplify our problems’ notation and will enable us
to develop tools to solve them.
• But before, remember what we did in the elimination method:

1) Interchange the ith and jth Equation

2) Multiply an equation by a nonzero constant
3) Replace the ith equation by c times the jth equation plus the ith Equation
for i≠j

These manipulations on a LES leads to an EQUIVALENT SYSTEM

MATRICES and
MATRIX OPERATIONS
Q. Why we are interested in matrices?
A Because
A. B we can express LES b by using
i ththem.

Q. Whyy we are interested in matrix operations?

p
A. They enable us to solve LES with a systematic
way

10
MATRIX: A matrix is rectangular
array of numbers denoted by
• If the matrix A has m rows and
n columns, we say that A is an
⎡ a 11 a 12 .. .. a 1n ⎤ m by n matrix (written mxn).
⎢a a 22 .. .. a 2n ⎥
⎢ 21 ⎥ • If m=n, we sayy that A is a
A = ⎢ .. .. .. .. .. ⎥ square matrix of order n and
⎢ ⎥ that the elements a11, a22,...,ann
⎢ .. .. .. .. .. ⎥
are on the main diagonal.
⎣⎢ a m 1 am2 .. .. a mn ⎥⎦

• The ith row of A is • We refer to aij as the (i,j) entry

[ai1 ai2......ain] 1≤i≤m (entry in the ith row and jth
column) or the (i,j)th element
and we write A=[aij]
• The jth column of A is ⎡ a 1 j ⎤
1≤j≤n ⎢a ⎥
⎢ 2j ⎥ • Amn indicates that A has m
⎢ .. ⎥ rows and n columns.
⎢ ⎥
⎢ .. ⎥ • If A is nxn we merely write An
⎢a mj ⎥
⎣ ⎦

• Example 1)
A is an 3 by 3 matrix (A3x3) ⎡1 2 3⎤
In matrix A a32=-3 A = 2 − 1 4⎥
⎢
⎢ ⎥
A is a square matrix of order 3 ⎢⎣0 − 3 2⎥⎦

B is an 1 by 3 matrix (B1x3
1 3) B = [1 3 − 7]
B is a row matrix
⎡2⎤
⎢− 1⎥
C is an 4 by 1 matrix (C4x1) C=⎢ ⎥
C is a column matrix ⎢3⎥
⎢ ⎥
⎣4⎦

D is an 2 by 2 matrix (D2x2) ⎡0 3⎤
D=⎢
D is a square matrix of order 2 ⎣ − 1 − 2⎥⎦

Q. What are the elements of A on

the main diagonal?
a11=?, a22=?, a33=?

11
• Definition 1.2 (Matrix Equality):
Two mxn matrices A=[aij] and B=[bij] are equal if they agree
entry by entry. That is, if aij= bij for i=1,2,...,m and j=1,2,...,n.
Example 2) If the matrices A and B are equal then find w, x, y
and z
⎡1 2 − 1⎤ ⎡1 2 w ⎤
A = ⎢⎢ 2 − 3 4 ⎥⎥ B = ⎢⎢ 2 x 4 ⎥⎥
⎢⎣0 − 4 5 ⎥⎦ ⎢⎣ y − 4 z ⎥⎦

The matrices A and B are equal then if and only if w=-1, x=-3, y=0 and
z=5

MATRIX OPERATIONS
• 1) Matrix Addition
• If A=[aij] and B=[bij] are both mxn matrices, then the sum A+B is an mxn
matrix C =[cij]
[cij] defined by cij=a
aij+ bij where i=1,2,...,m
i 1,2,...,m and j=1,2,...,n.
j 1,2,...,n.
In other words, to obtain the sum of A and B we add corresponding
entries.

• Example Let A and B are given matrices just below, Find matrix C=A+B.

⎡1 − 2 3 ⎤ ⎡0 2 1 ⎤
A=⎢ ⎥ B=⎢ ⎥
⎣ 2 − 1 4⎦ ⎣1 3 − 4 ⎦
⎡1+0 −2+2 3+1⎤ ⎡1 0 4⎤
A+B= ⎢ ⎥ =⎢ ⎥
⎣2+1 −1+3 4−4⎦ ⎣3 2 0⎦
• Note that: A and B are of the same size (i.e. They have same number of
rows and columns)

12
• 2. The Scalar Multiple of a Matrix
Given any matrix A and any number r. The matrix
r.A is obtained from the matrix A by multiplying
each element of A by r.
A=[aij]mxn and C=r.A
then cij=r.aij where i=1,2,...,m
, , , and j=1,2,...,n.
j , , ,
Example
⎡ −8 4 6 ⎤
⎡4 − 2 − 3⎤ and C=-2A, C = ⎢
A=⎢ ⎥ ⎣ − 14 6 − 4 ⎥⎦
⎣7 − 3 2⎦

Difference Between A and B

• If A and B are mxn matrices, we write
A ( 1) B as A
A+(-1).B A-BB and
d callll thi
this , “th
“the diff
difference
between A and B.
Example
⎡2 3 −5⎤ ⎡2 −1 3 ⎤
A= ⎢ ⎥ B= ⎢ ⎥
⎣4 2 1 ⎦ ⎣3 5 −2⎦

⎡2 − 2 3 +1 − 5 − 3⎤ ⎡0 4 − 8⎤
A−B = ⎢ ⎥=⎢ ⎥
⎣4 − 3 2 − 5 1+ 2 ⎦ ⎣1 − 3 3 ⎦

13
 LINEAR COMBINATION of
Matrices:
If A1, A2,.....Ak are mxn matrices and c1, c2,.., ck are real
numbers then an expression of the form
numbers,
k

c1.A1+c2.A2+...........+ck.Ak = ∑ cA
i =1
i i

is called a linear combination of A1, A2,.....Ak and c1,c2,..,ck are

called coefficients.
E
Example:
l L Lett us fifind
d th
the lilinear combination
bi ti b below:
l
⎡ 5 27 ⎤
− − 10
⎡0 − 3 5 ⎤ ⎡5 2 3⎤ ⎢ 2 2 ⎥
1 ⎢ 21 ⎥
3⎢⎢2 3 4 ⎥⎥ − ⎢⎢ 6 2 3⎥⎥ = ⎢ 3 8 ⎥
2 ⎢ 2 ⎥
⎢⎣1 − 2 − 3⎥⎦ ⎢⎣− 1 − 2 3⎥⎦ ⎢ 7 − 21⎥
−5
⎣⎢ 2 2 ⎥⎦

Matrix Multiplication
• Given two matrices A and B, the matrix product of A and
B ((written A.B)) is defined if and onlyy if the number of
columns in A is equal to the number of rows in B.
• If A=[aij] is an mxn matrix and B=[bij] is an nxp matrix,
then the product of A and B, A.B=C=[cij] is an mxp
matrix defined by
n
c ijj = ∑ a ik b kjj = a i1b1 j + a i 2 b 2 j + ....... + a in b njj
k =1

• where i=1,2,....,m and j=1,2,...,p

14
 ⎡ a11 a 12 ....... a1n ⎤
⎢a ....... a 2 n ⎥⎥ ⎡ b11 b12 ... b1 j ... b1p ⎤
⎢ 21 a 22 ⎢
⎢ .. .. ....... .. ⎥ ⎢b 21 b 22 ... b 2 j ... b 2 p ⎥⎥
A.B = ⎢ ⎥.
⎢ a i1 a i 2 ....... a in ⎥ ⎢ .. .. ... .. ... .. ⎥
⎢ .. ⎢ ⎥
.. ....... .. ⎥ ⎢⎣b n1 b n 2 ... b nj ... b np ⎦⎥
⎢ ⎥
⎢⎣a m1 a m 2 ....... a mn ⎦⎥

⎡ c11 c12 ... c1p ⎤

⎢c c 22 ... c 2 p ⎥⎥
=⎢
21
=C
⎢ .. .. c ij .. ⎥
⎢ ⎥
⎢⎣c m1 c m 2 ... c mp ⎦⎥

• Example

⎡− 2 5 ⎤
⎡1 2 − 1⎤
A = ⎢ B = ⎢⎢ 4 − 3 ⎥⎥
⎣3 1 4 ⎥⎦
⎢⎣ 2 1 ⎥⎦

⎡1(−2) + 2(4) + (−1)(2) 1(5) + 2(−3) + (−1)(1)⎤ ⎡4 − 2⎤

A.B = ⎢ ⎥ = ⎢6 16⎥
⎣ 3(−2) +1(4 ) + 4(2) 3(5) +1(−3 ) + 4(1) ⎦ ⎣ ⎦

15
• Examples
1)A2X3 and B3X4 then A.B2X4 and B.A is not defined
(why?)
2) A2X3 and B3X2 then A.B2X2 and B.A3x3
3) Let A and B are given below, find A.B and B.A,
show that A.B≠B.A
⎡ 1 2⎤ ⎡2 1⎤
A =⎢ ⎥ B=⎢
⎣− 1 3⎦ ⎣0 1⎥⎦
⎡ 1( 2 ) + 2 ( 0 ) 1(1) + 2 (1) ⎤ ⎡ 2 3⎤
A .B = ⎢ ⎥= ⎢− 2 2⎥
⎣ ( − 1)( 2 ) + 3 ( 0 ) ( − 1)(1) + 3 (1) ⎦ ⎣ ⎦
⎡2 1⎤ ⎡ 1 2⎤ ⎡ 1 7⎤
B.A = ⎢ ⎥.⎢ ⎥=⎢ ⎥
⎣0 1⎦ ⎣−1 3⎦ ⎣−1 3⎦

*The jth column of the matrix product A.B is a linear

combination of the columns of A with the
coefficient entries in Bj (the jth column of B)
Example: Show that the 2nd column of AB is a
linear combination of the columns of A with the
coefficient entries in B2 (the 2nd column of B)

⎡ 1 2⎤
⎡− 2 3 4⎤
A = ⎢⎢ 3 4⎥⎥ B= ⎢ ⎥
⎣ 3 2 1⎦
⎣⎢−1 5⎥⎦

⎡ 1 2⎤ ⎡ (1)(3) + (2)(2) ⎤ ⎡ 1 ⎤ ⎡2⎤

⎡3⎤ ⎢
AB2 = ⎢ 3 4⎥ ⎢ ⎥ = ⎢ (3)(3) + (4)(2) ⎥ = 3⎢⎢ 3 ⎥⎥ + 2 ⎢⎢4⎥⎥
⎢ ⎥ ⎥
2
⎢⎣− 1 5 ⎥⎦ ⎣ ⎦ ⎢⎣(−1)(3) + (5)(2)⎥⎦ ⎢⎣− 1⎥⎦ ⎢⎣5 ⎥⎦

16
 ⎡ 1 2⎤ ⎡ 1(−2) + 2(3) 1(3) + 2(2) 1(4) + 2(1) ⎤
⎢ ⎥ ⎡− 2 3 4⎤ ⎢
A.B = 3 4 .⎢ = 3(−2) + 4(3) 3(3) + 4(2) 3(4) + 4(1) ⎥
⎢ ⎥ ⎣ 3 2 1⎥⎦ ⎢ ⎥
⎣⎢− 1 5⎦⎥ ⎣⎢− 1(−2) + 5(3) − 1(3) + 5(2) − 1(4) + 5(1)⎦⎥

⎡ 4 7 6 ⎤
= ⎢⎢ 6 17 16 ⎥⎥
⎢⎣ 177 7 1 ⎥⎦

LINEAR SYSTEMS

a11x1 + a12x2 + .... + a1nxn = b1

a21x1 + a22x2 + .... + a2nxn = b2
# # + .... + # = # (2)
ai1x1 + ai2x2 + .... + ainxn = bi
# # + .... + # = #
am1x1 + am2x2 +....+
+ + amnxn = bm

17
 Let express this L.E.S by using matrix notation.
• The matrix A is called coefficient matrix.
• The matrix X is called as unknown matrix.
• The
Th matrix
t i B iis called
ll d R
R.H.S
H S matrix
ti

⎡ x1 ⎤
⎡ a11 a12 .. .. a1n ⎤ ⎢x ⎥ ⎡ b1 ⎤
⎢ a2 a .. .. a 2 n ⎥⎥ ⎢ 2⎥ ⎢b ⎥
⎢ 1 22
⎢ .. ⎥ ⎢ 2⎥
A = ⎢ .. .. .. .. .. ⎥ x=⎢ ⎥ B = ⎢ .. ⎥
⎢ ⎥ ⎢ .. ⎥ ⎢ ⎥
⎢ .. .. .. .. .. ⎥
⎢ .. ⎥ ⎢ .. ⎥
⎢⎣a m1 a m 2 .. .. a mn ⎦⎥ ⎢ ⎥ ⎣⎢b m ⎥⎦
⎣⎢ x n ⎦⎥

The Augmented Matrix

⎡ a 11 a 12 .. .. a 1n b1 ⎤
⎢ ⎥
⎢ a 21 a 22 .. .. a 2 n b 2 ⎥
⎢ .. .. .. .. .. .. ⎥
⎢ ⎥
⎢ .. .. .. .. .. .. ⎥
⎢ .. .. a mn b m ⎥⎦
⎣a m1 a m 2

• The coefficient matrix and augmented

g matrix
play key roles in our methods for solving
L.E.S.

18
 Matrix Representation of L.E.S
(Matrix Form of L.E.S)
• We can write AX=B, A is mxn and X is nx1, the
matrix product AX is mx1 matrix:

⎡ x1 ⎤
⎡ a11 a12 .. .. a1n ⎤ ⎢ ⎥ ⎡ a11 x1 + a12 x2 + ...... + a1n xn ⎤ ⎡ b1 ⎤
⎢a x2
a22 .. .. a2 n ⎥ ⎢ ⎥ ⎢ a21 x1 + a22 x2 + ...... + a2 n xn ⎥ ⎢b ⎥
⎢ 21 ⎥ ⎢ .. ⎥ ⎢ ⎥ ⎢ 2⎥
AX = ⎢ .. .. .. .. .. ⎥. ⎢ ⎥ = ⎢ ...................................... ⎥ = ⎢ .. ⎥
⎢ ..
⎢ .. .. .. .. .. ⎥⎥ ⎢ ⎥ ⎢⎢ .................................... ⎥⎥ ⎢ ⎥
⎢ .. ⎥
⎢ .. ⎥
⎢⎣am1 am 2 .. .. amn ⎥⎦ ⎢ ⎥ ⎢⎣am1 x1 + am 2 x2 + ...... + amn xn ⎥⎦ ⎢⎣bm ⎥⎦
⎣ xn ⎦

• Example Consider the following linear system.

3X1 +4X3-X4= 2
2X1 +X2 + X3 =-4
-X
X1+2X2+3X3+X4= 5
a)Write the system in matrix form.
b)Indicate the coefficient matrix of the linear
system.
c) Construct the augmented matrix of the system.
d) Write the linear system as a linear combination
of the columns of A.

19
a) The matrix form of the system

⎡ x1⎤
⎡ 3 0 4 − 1⎤ ⎢ ⎥ ⎡ 2 ⎤
⎢ 2 x2
⎢ 1 1 0 ⎥⎥ ⎢ ⎥ = ⎢⎢ − 4 ⎥⎥
⎢ x 3⎥
⎣⎢ − 1 2 3 1 ⎦⎥ ⎢ ⎥ ⎣⎢ 5 ⎦⎥
⎣x 4⎦

b)The coefficient matrix of the system

⎡ 3 0 4 − 1⎤
⎢2 1 1 0⎥
⎢ ⎥
⎢⎣− 1 2 3 1 ⎥⎦

c) The augmented matrix of the system

⎡ 3 0 4 −1 2 ⎤
⎢ ⎥
⎢ 2 1 1 0 − 4⎥
⎢⎣− 1 2 3 1 5 ⎥⎦

• Remember the matrix form of linear systems.

⎡ x1 ⎤
⎡ a11 a12 .. .. a1n ⎤ ⎢ ⎥ ⎡ a11 x1 + a12 x2 + ...... + a1n xn ⎤ ⎡ b1 ⎤
⎢a x2
a22 .. .. a2 n ⎢ ⎥ ⎢ a21 x1 + a22 x2 + ...... + a2 n xn ⎥
⎥ ⎢b ⎥
⎢ 21 ⎥ ⎢ .. ⎥ ⎢ ⎥ ⎢ 2⎥
AX = ⎢ .. .. .. .. .. ⎥. ⎢ ⎥ = ⎢ ...................................... ⎥ = ⎢ .. ⎥ = B
⎢ .. ⎥ .. ⎥ ⎢
⎢ .. .. .. .. ⎥ ⎢ .................................... ⎥⎥ ⎢ .. ⎥
⎢ .. ⎥ ⎢ ⎢ ⎥
⎣⎢am1 am 2 .. .. amn ⎥⎦ ⎢ ⎥ ⎢⎣am1 x1 + am 2 x2 + ...... + amn xn ⎥⎦ ⎣⎢bm ⎥⎦
⎣ xn ⎦

d) Representation of the linear system as a linear combination of the columns

of A where Ai is the ith column of A

⎡ a11 ⎤ ⎡
⎢ a 12 ⎥
⎤
⎡ a 1n ⎤ ⎡ b1 ⎤
⎢a ⎥ ⎢
⎢a
⎥
⎥ ⎢a ⎥ ⎢ b ⎥
n ⎢ 21 ⎥ ⎢ 22 ⎥ ⎢ 2n ⎥ ⎢ 2 ⎥
∑
⎢ ⎥
x i A i = x1 ⎢ .. ⎥ + x 2 ⎢⎢ .. ⎥⎥ + ...... + x n ⎢ .. ⎥ = ⎢ .. ⎥
i =1 ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ .. ⎥ ⎢ .. ⎥
⎢ ⎥ ⎢ .. ⎥ ⎢ .. ⎥
⎢⎣a m1 ⎥⎦ ⎢ ⎥
⎢⎣ a mn ⎥⎦ ⎣⎢a mn ⎥⎦ ⎣⎢b m ⎥⎦

20
 Definition (Transpose of a Matrix)
• If A=[aij] is an mxn matrix, then the transpose of A, AT=[aTij] is the nxm
matrix defined by aTij=aji.

• Thus the transpose of A is obtained from A by interchanging the rows and

columns of A.

• Example
⎡ 1 − 3⎤
⎡ 1 2 −1⎤
If A=⎢ ⎥ then A = ⎢⎢ 2 2 ⎥⎥
T

⎣− 3 2 7 ⎦ ⎢⎣−1 7 ⎥⎦

Algebraic Properties of Matrix

Operations
1) Properties of Matrix Addition
Let A,
A B and C be mxn matrices
a) A+B=B+A
b) A+(B+C)=(A+B)+C
c) There is a unique mxn matrix Omn such that A + Omn=A
for any mxn matrix. The matrix Omn is called the mxn
zero matrix. When m=n, we write On. (We can write
simply O for both cases)
d) For each mxn matrix A, There is a unique mxn matrix
D, such that A+D=O. We shall write D as –A so that
A+(-A)=0. The matrix –A is called the negative of A.

21
• Example 1)
⎡4 −1⎤ ⎡4 −1⎤ ⎡0 0⎤ ⎡4+0 −1+0⎤ ⎡4 −1⎤
If A = ⎢ ⎥ then ⎢ ⎥ +⎢ ⎥ = ⎢ ⎥ =⎢ ⎥
⎣2 3 ⎦ ⎣2 3 ⎦ ⎣0 0⎦ ⎣2+0 3+0 ⎦ ⎣2 3 ⎦

• Example 2)
⎡ 1 3 − 2⎤ ⎡− 1 − 3 2 ⎤
If A = ⎢ ⎥ , then − A = ⎢ ⎥
⎣− 2 4 3 ⎦ ⎣ 2 − 4 − 3⎦

Algebraic Properties of Matrix

Operations
• 2. Properties of Matrix Multiplication
a) If A, B and C are matrices, then
A.(B.C)=(A.B).C
b) If A, B and C are matrices, then
(A+B).C=A.C+B.C
c) If A, B and C are matrices, then
C.(A+B)=C.A+C.B

22
• Example 3
⎡1 0 2⎤
⎡2 −1 1 0⎤ ⎢2
⎡5 2 3⎤ −3 0 ⎥⎥
Let A = ⎢ , B = ⎢⎢ 0 2 2 2 ⎥⎥ and C = ⎢
⎣2 −3 4 ⎥⎦ ⎢0 0 3⎥
⎢⎣ 3 0 −1 3 ⎥⎦ ⎢ ⎥
⎣2 1 0⎦

⎡0 3 7⎤
⎡5 2 3⎤ ⎢ ⎡ 43 16 56 ⎤
Then A ( BC ) = ⎢ ⎥ .⎢8 − 4 6 ⎥⎥ = ⎢
⎣2 − 3 4⎦ 12 30 8 ⎥⎦
⎣⎢9 3 3 ⎦⎥ ⎣

⎡1 0 2⎤
⎡19 −1 6 13 ⎤ ⎢⎢ 2 − 3 0 ⎥⎥ ⎡ 43 16 56 ⎤
and ( AB ) C = ⎢ . =
⎣16 −8 − 8 6 ⎥⎦ ⎢ 0 0 3⎥ ⎢12 30
⎣ 8 ⎥⎦
⎢ ⎥
⎣2 1 0⎦

• Example 4
⎡1 0 ⎤
⎡ 2 2 3⎤ ⎡0 0 1 ⎤ ⎢ ⎥
Let A = ⎢ ⎥ , and B = ⎢2 3 − 1⎥ , and C = ⎢2 2 ⎥
⎣ 3 − 1 2⎦ ⎣ ⎦ ⎢⎣3 − 1⎥⎦

⎡1 0 ⎤
⎡ 2 2 4⎤ ⎢ ⎥ ⎡18 0⎤
Then (A + B)).C = ⎢ ⎥ .⎢2 2 ⎥ = ⎢12 3⎥
⎣ 5 2 1 ⎦ ⎢3 − 1⎥ ⎣ ⎦
⎣ ⎦

23
 Algebraic Properties of Matrix
Operations
• 3. Properties of Scalar Multiplication
• If r and s are real numbers and A and B
are matrices then
a) r.(s.A)=(r.s).A
b) (r+s)A=r.A+s.A
c) r.(A+B)=r.A+r.B
d) A.(r.B)=r.(A.B)=(r.A).B

• Example 5
⎡3 − 2 1 ⎤
⎡4 2 3⎤
Let A = ⎢ , and B = ⎢⎢ 2 0 − 1 ⎥⎥
⎣2 − 3 4 ⎥⎦
⎢⎣ 0 1 2 ⎥⎦

⎡12 6 9 ⎤ ⎡ 24 12 18 ⎤
Then 2 ( 3 A ) = 2 .⎢ = = 6A
⎣6 −9 12 ⎥⎦ ⎢⎣ 12 − 18 24 ⎥⎦

⎡6 −4 2 ⎤
⎡4 2 3⎤ ⎢ ⎡ 32 − 10 16 ⎤
Also A ( 2 .B ) = ⎢ . 4 0 − 2 ⎥⎥ = ⎢ = 2 .( AB )
⎣2 −3 4 ⎥⎦ ⎢ 0 0 26 ⎥⎦
⎣⎢ 0 2 4 ⎥⎦ ⎣

24
 Algebraic Properties of Matrix
Operations
• 4) Properties of Transpose

a) (AT)T=A
b) (A+B)T=AT+BT
c) (AB)T=BTAT
d) (rA)T=rAT

• Example 6
⎡1 − 2⎤
⎡ 1 2 3⎤
Let A=⎢ , then A T = ⎢⎢ 2 0 ⎥⎥
⎣− 2 0 1 ⎥⎦
⎢⎣ 3 1 ⎥⎦

⎡3 3⎤
⎡3 −1 2 ⎤
Let B=⎢ , then B T = ⎢⎢ − 1 2 ⎥⎥
⎣3 2 − 1⎥⎦
⎢⎣ 2 − 1⎥⎦

⎡4 1⎤
⎡ 4 1 5⎤ ⎢1 2 ⎥
Also A + B = ⎢ ⎥ , and ( A + B) T
= ⎢ ⎥
⎣ 1 2 0 ⎦ ⎢⎣5 0⎥⎦

⎡4 1⎤
Now A T
+ B T
= ⎢⎢ 1 2 ⎥⎥ = ( A + B ) T
⎢⎣ 5 0 ⎥⎦

25
• Example 7
⎡0 1⎤
⎡1 3 2⎤
Let A=⎢ , and B = ⎢⎢ 2 2 ⎥⎥
⎣2 − 1 3 ⎥⎦
⎢⎣ 3 − 1⎥⎦

⎡12 5 ⎤ ⎡12 7 ⎤
Then AB = ⎢ , and ( AB ) T = ⎢
⎣7 − 3 ⎥⎦ ⎣5 − 3 ⎥⎦

⎡1 2 ⎤
⎡0 2 3 ⎤
Also A T
= ⎢⎢ 3 − 1 ⎥⎥ , and BT = ⎢
⎣1 2 − 1 ⎥⎦
⎢⎣ 2 3 ⎥⎦

⎡ 12 7 ⎤
Then BTA T
= ⎢ = ( AB ) T
⎣ 5 − 3 ⎥⎦

Some important properties related

with matrix multiplication
• 1)AB need not equal BA
• 2)AB may be the zero matrix with A≠0 and
B≠0
• 3) AB may equal AC with B≠C

(These are also some differences between

matrix multiplication and the multiplication
of real numbers)

26
• Example 8
Consider the A and B matrices. Neither A nor B
is the zero matrix but AB=0
⎡1 2⎤ ⎡ 4 − 6⎤ ⎡0 0 ⎤
A=⎢ , B=⎢ , and AB = ⎢
⎣2 4 ⎥⎦ ⎣− 2 3 ⎥⎦ ⎥
⎣0 0 ⎦

• Example 9
Consider the following three matrices (A,B and C)
Although B≠C, AB=AC
⎡1 2⎤ ⎡2 1⎤ ⎡− 2 7 ⎤
A=⎢ ⎥, B= ⎢ ⎥ and C = ⎢ 5 −1⎥
⎣2 4⎦ ⎣3 2⎦ ⎣ ⎦
⎡8 5⎤
Then AB = AC = ⎢
⎣16 10 ⎥⎦

Special Types of Matrices

ZERO MATRIX: Omn
It is defined as the matrix all of whose entries are zero.

DIAGONAL MATRIX:
An nxn matrix A=[aij] is called a diagonal matrix if aij=0 for i≠j

SCALAR MATRIX:
It is a diagonal matrix whose diagonal elements are equal.

IDENTITY MATRIX:
The scalar matrix In=[aij] where aii=1 and aij=0 for i≠j is called nxn identity
matrix.

27
 Some rules related with special
matrices
• If A is any mxn matrix, then AIn=A and ImA=A

• If A is a scalar matrix, then A=rIn for some scalar r.

• If A is a square matrix and p is a positive integer then Ap=A.A....A (P factors)

• If A is nxn then A0=In

• If A is nxn then, for nonnegative integers p and q, APAq=A(p+q) and (Ap)q=Apq

• If A and B are square matrices and AB=BA then (AB)p=ApBp

Special Types of Matrices

UPPER TRIANGULAR MATRIX:
An nxn matrix A=[aijj] is called upper triangular if aijj=0 for i>j.

Example: The matrix A is upper triangular. ⎡1 2 3⎤

A = ⎢⎢ 0 3 5 ⎥⎥
⎢⎣ 0 0 2 ⎥⎦

LOWER TRIANGULAR MATRIX:

An nxn matrix A=[aij] is called lower triangular if aij=0 for i<j.

⎡1 0 0⎤
Example:The matrix B is lower triangular
B = ⎢⎢ 2 3 0 ⎥⎥
⎢⎣ 3 5 2 ⎥⎦

Note that: A diagonal matrix is both upper triangular and lower triangular

28
 Special Types of Matrices
SYMMETRIC MATRIX:
A matrix A is called symmetric
y if
AT=A
⎡1 2 3⎤ ⎡1 2 3⎤
Example:
A = ⎢⎢2 4 5⎥⎥, A = ⎢⎢2 4 5⎥⎥
T

⎣⎢3 5 6⎥⎦ ⎢⎣3 5 6⎦⎥

SKEW SYMMETRIC MATRIX:

A matrix A is called skew
symmetric if AT=-A
=A
⎡0 2 3 ⎤ ⎡0 − 2 − 3⎤
Example:
B = − 2 0 − 4⎥,
⎢ B = ⎢2 0
T
4⎥
⎢ ⎥ ⎢ ⎥
⎢⎣− 3 4 0 ⎥⎦ ⎢⎣3 − 4 0 ⎥⎦

Some observations about symmetric and skew symmetric

matrices

• If A is symmetric or skew symmetric then A is a square

matrix.
matrix
• If A is a symmetric matrix then the entries of A are
symmetric with respect to the main diagonal of A.
• A is symmetric if and only if aij=aji
• A is skew symmetric if and only if aij=-aji
• If A is skew symmetric, then the entries on the main
diagonal of A are zero.
• If A is an nxn matrix, then we can show that A=S+K
where S is symmetric and K is skew symmetric. This
decomposition is unique.

29
 Partitioned Matrices
• A matrix can be partitioned into submatrices by drawing horizontal
lines between rows and vertical lines between columns. The
partitioning can be carried out in many different ways
ways.
• Example
⎡a11 a12 a13 a14 a15 ⎤ ⎡a11 a12 a13 a14 a15 ⎤
⎢a a22 a23 a24 a25 ⎥ ⎢a a22 a23 a24 a25 ⎥
A = ⎢ 21 ⎥, A = ⎢ 21 ⎥,
⎢a31 a32 a33 a34 a35 ⎥ ⎢a31 a32 a33 a34 a35 ⎥
⎢ ⎥ ⎢ ⎥
⎣a41 a42 a43 a44 a45 ⎦ ⎣a41 a42 a43 a44 a45 ⎦

⎡A A12 ⎤ ⎡ Aˆ11 Aˆ12 Aˆ13 ⎤

A = ⎢ 11 , A=⎢ ⎥
⎣ A21 A22 ⎥⎦ ⎢⎣ Aˆ 21 Aˆ 22 Aˆ 23 ⎥⎦
• NOTE THAT: The augmented matrix of a LES is a partitioned
matrix. Thus If AX=B, we can write the augmented matrix of this
system as [A|B]

Block Multiplication
• The method of multiplying partitioned matrices is known as block
multiplication.
• If A is partitioned as shown previously and B is given and
partitioned
ii d jjust as b
below,
l then
h AB iis d
defined
fi d as ffollows:
ll
⎡ b 11 b 12 b 13 b 14 ⎤
⎢b b 22 b 23 b 24 ⎥⎥ ⎡ B 11 B 12 ⎤
⎢ 21
B = ⎢ b 31 b 32 b 33 b 34 ⎥ = ⎢⎢ B 21 B 22 ⎥⎥
⎢ ⎥
⎢ b 41 b 42 b 43 b 44 ⎥ ⎢⎣ B 31 B 32 ⎥⎦
⎢⎣ b 51 b 52 b 53 b 54 ⎥⎦

⎡ Â B + Â12 B21 + Â13 B31 Â11B12 + Â12 B22 + Â13 B32 ⎤

AB = ⎢ 11 11 ⎥
⎣Â 21B11 + Â 22 B21 + Â 23 B31 Â 21B12 + Â 22 B22 + Â 23 B32 ⎦

30
 ⎡ B11 B12 ⎤
⎡A
 A12
 A13 ⎤
 B = ⎢⎢B21 B 22 ⎥⎥
A = ⎢ 11 ⎥
Â
⎣ 21 Â 22 Â 23 ⎦ ⎣⎢ B31 B32 ⎥⎦

⎡ Â B + Â12 B21 + Â13B31 Â11B12 + Â12 B22 + Â13 B32 ⎤

AB = ⎢ 11 11 ⎥
⎣Â 21B11 + Â 22 B21 + Â 23B31 Â 21B12 + Â 22 B22 + Â 23B32 ⎦

• Example
⎡1 0 1 0⎤
⎢0 2 3 −1⎥⎥ ⎡A11 A12 ⎤
Let A = ⎢ =
⎢2 0 − 4 0 ⎥ ⎢⎣A21 A22 ⎥⎦
⎢ ⎥
⎣0 1 0 3⎦

⎡ 2 0 0 1 1 − 1⎤
⎢ 0 1 1 −1 2 2 ⎥⎥ ⎡ B 11 B 12 ⎤
and B=⎢ =
⎢ 1 3 0 0 1 0 ⎥ ⎢⎣ B 21 B 22 ⎥⎦
⎢ ⎥
⎣− 3 −1 2 2 0 − 1⎦

⎡3 3 0 1 2 −1 ⎤
⎢6 12 0 −3 7 5 ⎥⎥ ⎡ C11 C12 ⎤
Then AB = C = ⎢ =
⎢ 0 − 12 0 2 − 2 − 2⎥ ⎢⎣C 21 C 22 ⎥⎦
⎢ ⎥
⎣− 9 − 2 7 2 2 − 1⎦

31
• Where C11 should be A11B11+A12B21. We verify
that C11 in this expression as follows:

⎡1 0⎤ ⎡2 0 0⎤ ⎡1 0 ⎤ ⎡ 1 3 0⎤
A11B11 + A12 B21 = ⎢ ⎥.⎢ ⎥ +⎢ ⎥.⎢ ⎥
⎣0 2⎦ ⎣0 1 1⎦ ⎣3 − 1⎦ ⎣− 3 − 1 2⎦

⎡2 0 0⎤ ⎡1 3 0 ⎤ ⎡3 3 0⎤
=⎢ ⎥ +⎢ ⎥=⎢ ⎥ = C11
⎣0 2 2⎦ ⎣6 10 − 2⎦ ⎣6 12 0⎦

Benefit of partitioning
• This method of multiplying partitioned matrices is also
known as block multiplication.

• Partitioned matrices can be used to great advantage in

dealing with matrices that exceed the memory capacity of a
computer. Thus, in multiplying two partitioned matrices, one
can keep the matrices on disk and only bring into memory
the submatrices required to form submatrix products. The
latter, of course, can be put on disk as they are formed.

• The partitioning must be done so that the products of

corresponding submatrices are defined. In contemporary
computing technology, parallel processing computers use
partitioned matrices to perform matrix computations more
rapidly.

32
 Nonsingular Matrices
• Definition: An nxn matrix A is called
nonsingular or invertible
nonsingular, invertible, if there exist an nxn
matrix B such that AB=BA=In
• Such a B is called an inverse of A.
• Otherwise, A is called singular, or noninvertible
• Example
⎡2 3⎤ ⎡− 1 3 / 2⎤
L
Let A=⎢ ⎥ , andd B = ⎢ ⎥
⎣2 2⎦ ⎣ 1 −1 ⎦

• Since AB=BA=I2 we conclude that B is an

inverse of A.

Theorem 1.5
The inverse of a matrix, if it exists, is unique.

QUESTION: If the A matrix is given as follows,

show that whether A is a nonsingular matrix or
not.
ANSWER: if A-1 exists,then let us assume the
inverse matrix as follows:
⎡ 1 2⎤ ⎡a b ⎤
A=⎢ ⎥, let A−1 = ⎢ ⎥
⎣2 4 ⎦ ⎣c d ⎦

33
 ⎡ 1 2⎤ ⎡a b ⎤ ⎡ 1 0⎤
Then we must have A.A −1 = ⎢ ⎥.⎢ ⎥ = I2 = ⎢ ⎥
⎣2 4⎦ ⎣c d ⎦ ⎣0 1⎦
⎡ a + 2c b + 2d ⎤ ⎡ 1 0⎤
so that ⎢ ⎥=⎢ ⎥
⎣2a + 4c 2b + 4d ⎦ ⎣0 1⎦
Equating corresponding entries of two matrices,
we obtain the following LES
a + 2c = 1 b + 2d = 0
2a + 4c = 0 2b + 4d = 1
This LES has no solution (inconsistent system).
Thus A is singular
(Our assumption A −1 exists is incorrect.)

Example: If the A matrix is given as follows and if

A-1 exists, let us show that whether A is a
nonsingular matrix or not.

⎡1 2 ⎤ ⎡a b⎤
A = ⎢ ⎥ and let A -1 = ⎢ ,
⎣3 4 ⎦ ⎣c d ⎥⎦
then we must have A.A -1 = I 2
−1 ⎡1 2 ⎤ ⎡ a b⎤ ⎡1 0⎤
A .A = ⎢ .
⎥ ⎢ = I = ⎢0 1 ⎥,
d ⎥⎦
2
⎣3 4 ⎦ ⎣ c ⎣ ⎦
⎡ a + 2c b + 2d ⎤ ⎡1 0⎤
so that ⎢ ⎥ = ⎢0 1 ⎥
⎣3a + 4 c 3b + 4 d ⎦ ⎣ ⎦

34
-3/ a+2c=1 -3/ b+2d=0
3a+4c=0 3b+4d=1
-3a-6c=-3 -3b-6d=0
3a+4c=0 3b+4d=1
-2c=-3 -2d=1
c=3/2 d=-1/2
a+2(3/2)=1 b+2(-1/2)
a=-2 b=1

• We conclude that A is nonsingular and A-1 is found as follows:

⎡a b ⎤ ⎡ − 2 1 ⎤
A −1 = ⎢ ⎥ =⎢ ⎥
⎣ c d ⎦ ⎣3 / 2 − 1 / 2⎦

SOME PROPERTIES of INVERSES of

MATRICES
Theorem 1.6
If A and B are both nonsingular nxn
matrices, then AB is nonsingular and
(AB)-1=B-1A-1
Corollary 1.1
If A 1,A2,.....,Ar are nxn nonsingular matrices, then A 1.A2.......Ar is
nonsingular and
(A 1.A2.......Ar)-1=A-1r.A-1r-1.......A-11
Theorem 1.7
matrix then A-11 is nonsingular and
If A is a nonsingular matrix,
-1 -1
(A ) =A
Theorem 1.8
If A is a nonsingular matrix, then AT is nonsingular and (A-1)T=(AT)-1

35
•Example
In our previous example we found the inverse of A
when A is given as follows:

⎡1 2⎤ ⎡−2 1 ⎤
A=⎢ , A -1 = ⎢
⎣3 4 ⎥⎦ ⎣3 / 2 − 1 / 2 ⎥⎦
⎡- 2 3/2 ⎤
( A −1 )T = ⎢ ⎥
⎣ 1 - 1/2 ⎦

⎡1 3⎤ ⎡− 2 3 2⎤
AT = ⎢ ⎥ and ( AT )−1 = ⎢
⎣2 4 ⎦ ⎣ 1 − 1 2 ⎥⎦

LINEAR SYSTEMS and

INVERSES
If A is an nxn matrix, then the linear system AX=B
y
is a system of n equations
q in n unknowns.
Suppose that A is nonsingular, then A-1 exists and
we can multiply AX=B by A-1on both sides,
obtaining
A (AX)= A-1 B
-1

InX= A-1 B
X= A-1 B
That is clearly solution to the given LES.
Thus, if A is nonsingular, we have a unique
solution

36
• Example
⎡1 2⎤
Lets consider A = ⎢ and
⎣3 4 ⎥⎦
⎡-2 1 ⎤
its inverse matrix is calculated as A -1 = ⎢
⎣ 3/2 - 1/2 ⎥⎦
⎡8 ⎤
If B = ⎢ ⎥ then the solution t o the LES AX = B is
⎣6⎦
⎡−2 1 ⎤ ⎡ 8 ⎤ ⎡ − 10 ⎤
X = ⎢ ⎥ .⎢ ⎥ = ⎢ ⎥
⎣3 / 2 − 1 / 2 ⎦ ⎣6 ⎦ ⎣ 9 ⎦

ECHELON FORM of a MATRIX

Definition: An mxn matrix A is said to be in reduced row
echelon form if it is satisfies the following
gpproperties:
p
a) All zero rows; if there are any, appear at the bottom of
the matrix.
b) The first nonzero entry from the left of a nonzero row is
a 1. This entry is called a leading one of its row.
c) For each nonzero row, the leading one appears to the
right and below of any leading one’s in preceding rows.
d) If a column contain a leading one, then all other entries
in that column are zero.
Definition: An mxn matrix satisfying properties (a), (b) and
(c) is said to be in row echelon form.

37
• Example
The following matrices are in reduced row echelon
form since they satisfy properties (a),(b),(c) and
(d).
⎡1 0 0 0 −2 4 ⎤
⎡1 0 0 0⎤ ⎢0
⎢0 1 0 0 4 8 ⎥⎥
1 0 0 ⎥⎥ ⎢
A=⎢ , B = ⎢0 0 0 1 7 − 2⎥
⎢0 0 1 0⎥ ⎢ ⎥
⎢ ⎥ ⎢0 0 0 0 0 0 ⎥
⎣0 0 0 1⎦
⎢⎣ 0 0 0 0 0 0 ⎥⎦
⎡1 2 0 0 1⎤
C = ⎢⎢ 0 0 1 2 3 ⎥⎥
⎢⎣ 0 0 0 0 0 ⎥⎦

• Example
The following matrices are not in reduced row
echelon form.(why not?)

⎡1 2 0 4 ⎤ ⎡1 0 3 4 ⎤
D = ⎢⎢0 0 0 0 ⎥⎥, E = ⎢⎢0 2 - 2 5⎥⎥
⎣⎢0 0 1 − 3⎦⎥ ⎣⎢0 0 1 2⎥⎦

⎡1 0 3 4⎤ ⎡1 2 3 4⎤
⎢0 1 - 2 5⎥⎥ ⎢0 1 - 2 5⎥⎥
F= ⎢ , G= ⎢
⎢0 1 2 2⎥ ⎢0 0 1 2⎥
⎢ ⎥ ⎢ ⎥
⎣0 0 0 0⎦ ⎣0 0 0 0⎦

38
 We shall now show that every matrix can be put into row
echelon form or into reduced row echolon form, by means
of certain row operations
Definition 1.11: ELEMENTARY ROW(COLUMN) OPERATIONS
An elementary row(column) operation on a matrix A is any one of the
following operations:
(a) Type I: Interchange any two rows (columns)
(b) Type II: Multiply a row (column) by a nonzero number
(c) Type III: Add a multiple of one row (column) to another.

REMEMBER that we have talked about these operations as

manipulations in the elimination method. After applying these
manipulations to the LES we had equivalent linear systems.

When a matrix is viewed as the augmented matrix of a linear system,

after appliying these elementary row operations to the matrix we
will get the row equivalent matrix.

• Definition 1.12
• An mxn matrix B is said to be row equivalent to an mxn matrix A, if
B can be obtained by applying a finite sequence of elementary row
operations to A.
• B is row equivalent to A, C is row equivalent to B, D is row
equivalent to C.
C Then it follows that D is row equivalent to A,
A since
we obtained D by applying three elementary row operations to A.

⎡1 2 4⎤ 2R +R ⎡1 2 4⎤ R ↔R ⎡1 2 4⎤
A = ⎢⎢2 1 3⎥⎥ → ⎢ ⎥
B = ⎢4 - 3 7⎥ → C = ⎢⎢1 - 2 2⎥⎥
3 2 2 3

⎢⎣1 − 2 2⎥⎦ ⎢⎣1 - 2 2⎥⎦ ⎢⎣4 - 3 7⎥⎦

⎡1 2 4⎤ 2R ⎡2 4 8⎤
⎢ ⎥
C = ⎢1 - 2 2⎥ → D = ⎢⎢1 - 2 2⎥⎥
1

⎢⎣4 - 3 7⎥⎦ ⎢⎣4 - 3 7⎥⎦

39
SOME PROPERTIES
1) Every matrix is row equivalent to itself.
2) If B is row equivalent to A, then A is row
equivalent to B
B. We can say A and B are row
equivalent.
3) If C is row equivalent to B and B is row
equivalent to A, then C is row equivalent to A.

THEOREM 1.9
Every nonzero matrix A=[aij] is row equivalent
to a matrix in row echelon form

Example
Fi d th
Find the row echelon
h l
form of the matrix A. ⎡0 2 − 4⎤
3
⎢0 0 3 ⎥⎥
2
A=⎢
⎢2 2 −5 2 ⎥
⎢ ⎥
⎣2 0 −6 9 ⎦

40
⎡0 2 3− 4⎤ ⎡2 2 −5 2⎤ ⎡1 1 −5/ 2 1⎤
⎢0 0 2 ⎥
3 ⎥ 3 1 ⎢⎢0
R ↔ R 0 2 ⎥
3 ⎥ (1/ 2) R1 ⎢⎢0 0 2 3 ⎥⎥ − 2 R 2 + R 4
⎢ → → →
⎢2 2 −5 2 ⎥ ⎢0 2 3 − 4⎥ ⎢0 2 3 − 4⎥ R 4
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣2 0 −6 9 ⎦ ⎣2 0 −6 9 ⎦ ⎣2 0 −6 9⎦

⎡1 1 − 5 / 2 1 ⎤ ⎡1 1 − 5 / 2 1 ⎤ ⎡1 1 − 5 / 2 1 ⎤
⎢0 0 2 3⎥⎥ R ↔ R ⎢ 0 2 3 ⎥
− 4⎥ (1 / 2 ) R ⎢0 1 3 / 2 − 2⎥⎥ 2 R 2 + R 4
⎢ → ⎢ → ⎢
3 2 2
→
⎢0 2 3 − 4⎥ ⎢0 0 2 3⎥ ⎢0 0 2 3 ⎥ R4
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣0 − 2 −1 7⎦ ⎣0 − 2 −1 7⎦ ⎣0 − 2 −1 7⎦

⎡1 1 −5/ 2 1⎤ ⎡1 1 −5/ 2 1 ⎤ ⎡1 1 −5/ 2 1 ⎤

⎢0 1 3/ 2 − 2⎥⎥ (1/ 2) R 3 ⎢⎢0 1 3/ 2 − 2 ⎥⎥ − 2 R 3 + R 4 ⎢⎢0 1 3/ 2 − 2 ⎥⎥
⎢ → →
⎢0 0 2 3⎥ ⎢0 0 1 3 / 2 ⎥ R 4 ⎢0 0 1 3 / 2⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣0 0 2 3⎦ ⎣0 0 2 3 ⎦ ⎣0 0 0 0 ⎦

• The last matrix (say H) is in row echelon

f
form and
d it is
i row equivalaent
i l t tto A

Theorem 1.10
Every nonzero matrix A=[aij] is row
equivalent to a matrix in reduced row
echelon form

41
Example
Fi d a matrix
Find t i iin reduced
d d
row echelon form that
is row equivalent to
the matrix A. ⎡0 2 3 − 4⎤
⎢0 0 2 3 ⎥⎥
A = ⎢
⎢2 2 −5 2 ⎥
⎢ ⎥
⎣2 0 −6 9 ⎦

• We start with matrix H which is row equivalent to A. Last

matrix (say K) is in reduced row echelon form and is row
equivalent to A.
⎡1 1 −5/ 2 1 ⎤ ⎡1 1 − 5 / 2 1 ⎤
⎢0 1 ⎥ ⎢
3 / 2 − 2 ⎥ ( −3 / 2) R 3 +R 2 ⎢0 1 0 −17 / 4⎥⎥ (5 / 2) R 3 +R1
⎢ → →
⎢0 0 1 3 / 2⎥ R 2 ⎢0 0 1 3 / 2 ⎥ R1
⎢ ⎥ ⎢ ⎥
⎣0 0 0 0 ⎦ ⎣0 0 0 0 ⎦
⎡1 1 0 19 / 4 ⎤ ⎡1 0 0 9 ⎤
⎢0 1 0 −17 / 4⎥⎥ −R 2 +R1 ⎢⎢0 1 0 −17 / 4⎥⎥
⎢ →
⎢0 0 1 3 / 2 ⎥ R1 ⎢0 0 1 3 / 2 ⎥
⎢ ⎥ ⎢ ⎥
⎣0 0 0 0 ⎦ ⎣0 0 0 0 ⎦

42
 We now apply these results to the
solution to the LES
Theorem 1.11
Let AX=B
AX B and CX=D
CX D be two linear equation
systems, each of m equations in n unknowns.If
the augmented matrices [A|B] and [C|D] are row
equivalent, then the linear systems are
equivalent, that is, they have same solutions.
Corollary 1.2
If A and C are row equivalent
q mxn matrices,, then
the homogeneous systems AX=0 and CX=0 are
equivalent.

SOLVING LINEAR SYSTEMS

• We now have esential features for solving linear
systems.
y

• The idea consists of starting with the linear system

AX=B, then obtaining a partitioned matrix [C|D] in either
row echelon form or reduced row echelon form, that is
row equivalent to the augmented matrix [A|B].

• Now [C|D] represents the linear system CX=D, which is

quite simply to solve because of the structure of [C|D],
and the set of solutions to the system gives precisely the
set of solutions to AX=B.

43
 SOLVING LINEAR SYSTEMS
• The method where [C|D] is in row echelon
f
form is
i called
ll d GAUSSIAN ELIMINATION
ELIMINATION.

• The method where [C|D] is in reduced row

echelon form is called GAUSS-JORDAN
REDUCTION.
REDUCTION

GAUSSIAN ELIMINATION
METHOD
STEP 1.
Th ttransformation
The f ti off th
the augmentedt d
matrix [A|B] to the matrix [C|D] in row
echelon form using elementary row
operations.
STEP 2.
Solution of the linear system
corresponding to the augmented matrix
[C|D] using back substitution.

44
• For the case in which
A is nxn
nxn, and the
linear system AX=B
⎡1 c12 c13 .. .. c1n d1 ⎤
has a unique solution, ⎢ ⎥
⎢0 1 c 23 .. .. c2n d2 ⎥
the matrix [C|D] has
⎢.. .. .. .. .. .. .. ⎥
the following form ⎢ ⎥
⎢0 0 0 .. 1 c ( n −1) n d n −1 ⎥
⎢0 0 0 .. 0 1 d n ⎥⎦
⎣

• The augmented matrix represents the linear

system
x 1 + c 12 x 2 + c 13 x 3 + .......... ... + c 1 n x n = d 1
x 2 + c 23 x 3 + .......... ... + c 2 n x n = d 2
.......... ...
.......... ...
x (n -1) + c (n -1)n x n = d n -1
x n = dn

45
• Back substitution proceeds from the nth equation
upward solving one variable from each
upward,
equation:
xn = d n
xn −1 = d ( n −1) − c( n−1) n xn
....
x2 = d 2 − c23 x3 − c24 x4 − ....... − c2 n xn
x1 = d1 − c12 x2 − c13 x3 − ............ − c1n xn

• Example
S l the
Solve th following
f ll i lilinear system
t b
by using
i
Gaussian Elimination Method

x1 + 2x 2 + 3x 3 = 9 ⎡ 1 2 3 9⎤
2x 1 - x 2 + x 3 = 8 [A B ] = ⎢⎢2 − 1 1 8⎥⎥
3x1 - x3 = 3 ⎢⎣3 0 − 1 3⎥⎦

46
• Transforming the augmented matrix to row
echelon form we obtain,

⎡1 2 3 9⎤ ⎡1 2 3 9 ⎤
−2R +R (−1/5)R
⎢ ⎥ 1 2
⎢ ⎥ 2

⎢2 −1 1 8⎥ → ⎢0 − 5 − 5 − 10 ⎥ →
R
⎢⎣ 3 − 1 3 ⎥⎦ ⎣⎢ 3 ⎥⎦
2
0 0 −1 3

⎡1 2 3 9⎤ ⎡1 2 3 9⎤
−3R +R +6R
⎢ ⎥ 1 3
⎢ ⎥ R 3 2

⎢0 1 1 2⎥ → ⎢0 1 1 2
⎥ →
R R

⎣⎢ 3 − 1 3 ⎥⎦ ⎢⎣ 0 − 10 − 24 ⎥⎦
3 3
0 − 6

⎡1 2 3 9 ⎤ ⎡1 2 3 9⎤
(−1/ 4)R
⎢ ⎥ 3
⎢ ⎥
⎢0 1 1 2 ⎥ → ⎢0 1 1 2⎥
⎢⎣ 0 0 − 4 − 12 ⎥⎦ ⎢⎣ 0 0 1 3 ⎥⎦

⎡1 2 3 9⎤
We obtain [C D ] = ⎢⎢0 1 1 2⎥⎥
⎢⎣0 0 1 3⎥⎦
⎡ 1 2 3⎤ ⎡ x1 ⎤ ⎡9⎤
⎢0 1 1⎥ .⎢ x ⎥ = ⎢2⎥
⎢ ⎥ ⎢ 2⎥ ⎢ ⎥
⎢⎣0 0 1⎥⎦ ⎢⎣ x3 ⎥⎦ ⎢⎣3⎥⎦
x1 + 2x 2 + 3x 3 = 9
x2 + x3 = 2
x3 = 3
Using
U i back
b k substituti
b tit tion, we have
h
x3 = 3
x2 + 3 = 2 ⇒ x 2 = -1
x1 + 2(-1) + 3(3) = 9 ⇒ x1 = 9 - 2(-1) - 3(3) = 2

47
• For the general case in which A is mxn is
h dl d iin a similar
handled i il ffashion,
hi
The linear system AX=B, the matrix [C|D] is
the row echelon form of the augmented
matrix of the linear system.
Thus we consider CX=D
CX=D, where C is mxn
mxn,
[C|D] might be of the following form:

⎡1 c12 c13 c1n d1 ⎤

⎢ ⎥
⎢0 0 1 c 24 c2n d2 ⎥
⎢.. .. .. .. ...... .. .. .. ⎥
⎢ ⎥
⎢.. .. .. .. ...... .. .. .. ⎥
⎢0 0 0 0 ...... 1 c ( k −1) n d ( k −1) ⎥
⎢ ⎥
⎢0 0 0 0 ....... 0 1 dk ⎥
⎢.. .. .. .. ....... 0 0 d k +1 ⎥
⎢ ⎥
⎢⎣0 0 0 0 ....... 0 0 d m ⎥⎦

48
• This augmented matrix represents the following
linear system:
x1 + c12 x2 + c13 x3 + .................... + c1n xn = d1
x3 + c24 x4 + ......... + c2 n xn = d 2
# # #
x( n −1) + c( k −1) n xn = d ( k −1)
xn = d k
0 x1 + ................................................. + 0 xn = d ( k +1)
# # #
0 x1 + .................................................. + 0 xn = d m

Some Remarks

1)If dk+1=1 then CX=D has no solution, since at least one

equation is not satisfied.

2)If dk+1=0 (Which implies that dk+2=........=dm =0) then we

use the back substitution.

a) If in the solution some of the unknowns are

expressed in terms of others this indicates
that CX=D has infinitely many solutions
solutions.

b) If every unknown have determined value

this indicate a unique solution.

49
• Example
⎡1 2 3 4 5 6⎤
⎢0 1 2 3 − 1 7⎥
[C D] = ⎢ ⎥
⎢0 0 1 2 3 7⎥
⎢ ⎥
⎣0 0 0 1 2 9⎦
By using back substitution,
x 4 + 2x 5 = 9 we can give any real number to
an unknown which is represented by a column
without a leading one.
x 5 = any real number, say r
x 4 = 9 - 2x 5 ⇒ x4 = 9 − 2r
x 3 = 7 - 2x 4 - 3x 5 ⇒ x3 = 7 − 2(9 − 2r ) − 3r = −11 + r
x2 = 7 − 2 x3 − 3 x4 + x5 ⇒ x2 = 7 - (-11 + r) - 3(9 - 2r) + r = 2 + 5r
x1 = 6 − 2 x2 − 3x3 − 4 x4 − 5 x5 ⇒ x1 = -1 - 10r
The given linear system has infinitely many solutions

GAUSS-JORDAN REDUCTION
METHOD
• When using the Gauss-Jordan Reduction
P
Procedure,
d we ttransform
f the
th augmented t d
matrix [A|B] to [C|D] which is in reduced
echelon form.
• This means that we can solve linear
system CX=D
CX D without back substitution.

50
 • Example
• Solve
S l ththe lilinear system
t b
below
l b
both
th b
by G
Gaussian
i
Elimination Method and Gauss Jordan
Reduction Method.

• x1+2x2+3x3=6
• 2x1-3x2+2x3=14
• 3x1+x2-x3=-2

⎡1 2 3 6⎤ ⎡1 2 3 6⎤ ⎡1 2 3 6 ⎤
⎢ ⎥ − 2 R1 + R 2 ⎢ ⎥ −3 R 1 + R 3 ⎢ ⎥ 2 ↔R 3
⎢2 − 3 2 14 ⎥ ⎯⎯R⎯ ⎯→ ⎢0 − 7 − 4 2 ⎥ ⎯⎯ ⎯
R
⎯→ ⎢0 − 7 − 4 2 ⎥ ⎯R⎯ ⎯⎯→
( −1 / 5 ) R 3
⎢⎣3 1 − 1 − 2⎥⎦ ⎢⎣3 1 − 1 − 2⎥⎦ ⎢⎣0 − 5 − 10 − 20⎥⎦
2 3

⎡1 2 3 6⎤ ⎡1 2 3 6 ⎤ ⎡1 2 3 6 ⎤
⎢ ⎥ 7R 2 +R 3 ⎢ ⎥ (1/10 ) R 3 ⎢ ⎥
⎢0 1 2 4⎥ ⎯⎯ ⎯ ⎯→ ⎢0 1 2 4 ⎥ ⎯⎯ ⎯ ⎯→ ⎢0 1 2 4⎥
R3
⎢⎣0 − 7 − 4 2⎥⎦ ⎢⎣0 0 10 30⎥⎦ ⎢⎣0 0 1 3⎥⎦
The last matrix is in row echelon form. Using back sustitutio n
x3 = 3
x2 + 2x3 = 4
x2 + 2(3) = 4
x2 = -22
x1 + 2x2 + 3x3 = 6
x1 + 2(-2) + 3(3) = 6
x1 - 4 + 9 = 6
x1 = 1

51
• We found the solution by Gaussian
Eli i ti
Elimination.
• To solve the same linear system by
Gauss-Jordan Reduction, we transform
the last matrix which is in row echelon
form to the reduced row echelon form.

The Solution by Gauss-Jordan

Reduction Method
⎡1 2 3 6⎤ ⎡1 2 3 6 ⎤
⎢ ⎥ −2R 3 +R 2 ⎢ ⎥
⎢0 1 2 4 ⎥ ⎯ ⎯ ⎯⎯ → ⎢ 0 1 0 − 2 ⎥ ⎯ −⎯
3R 3 + R1
⎯⎯ →
R2 R1
⎢⎣ 0 0 1 3 ⎥⎦ ⎣⎢ 0 0 1 3 ⎥⎦
⎡1 2 0 − 3⎤ ⎡1 0 0 1 ⎤
⎢ ⎥ −2R 2 +R1 ⎢ ⎥
⎢0 1 0 − 2 ⎥ ⎯ ⎯ ⎯⎯ → ⎢ 0 1 0 − 2⎥
R1
⎢⎣ 0 0 1 3 ⎥⎦ ⎢⎣ 0 0 1 3 ⎥⎦
The solution is
x1 = 1
x 2 = -2
x 3 = 3, as before.

52
• Remember the definition of homogeneous
systems
t
• Remember the definition of the trivial
solution
• Remember the definition of the nontrivial
solution

• Theorem 1.12
A homogeneous system of m linear
equations in n unknowns always has a
nontrivial solution if m<n, that is,
if the number of unknowns exceeds the
number of equations.
equations

53
• Example
• Consider the homogeneous system
x1+x2+x3+x4=0
x1+x4=0
x1+2x2+x3=0
The augmented matrix of the system

⎡1 1 1 1 0⎤
⎢ ⎥
⎢1 0 0 1 0⎥
⎣⎢1 2 1 0 0 ⎥⎦

⎡1 1 1 1 0⎤ ⎡1 1 1 1 0⎤
⎢ ⎥ ⎢ ⎥
⎢1 0 0 1 0 ⎥ ⎯ R⎯1 −⎯
R 2
→ ⎢0 1 1 0 0 ⎥ ⎯ R⎯1 −⎯
R3
→
R 2 R3
⎢⎣ 1 2 1 0 0 ⎥⎦ ⎢⎣ 1 2 1 0 0 ⎥⎦
⎡1 1 1 1 0⎤ ⎡1 1 1 1 0⎤
⎢ ⎥ ⎢ ⎥
⎢0 1 1 0 0 ⎥ ⎯ R⎯2 +⎯
R3
→ ⎢0 1 1 0 0 ⎥ ⎯ R⎯1 −⎯
R 2
→
R3 R1
⎢⎣ 0 −1 0 1 0⎦⎥ ⎢⎣ 0 0 1 1 0⎦⎥
⎡1 0 0 1 0 ⎤ ⎡1 0 0 1 0⎤
⎢ ⎥ R 2−R 3 ⎢ ⎥
⎢ 0 1 1 0 0 ⎥ ⎯ ⎯R ⎯ → ⎢0 1 0 − 1 0 ⎥
⎢⎣ 0 0 1 1 0 ⎥⎦ ⎢⎣ 0 0 1 1 0 ⎥⎦
2

Hence , the solution is found by the following equations

x1+x4=0
x2-x4=0
x3+x4=0,
we can give any real number to the unknown which is represented by
the column without a leading one. So, x4=r, any real number, x1=-r,
x2=r, x3=-r (infinitely many solutions)

54
• Example
Consider the homogeneous system whose
augmented matrix is as follows:
⎡1 0 0 0 2 0 ⎤
⎢ ⎥
⎢0 0 1 0 3 0 ⎥
⎢0 0 0 1 4 0 ⎥
⎢ ⎥
⎣⎢ 0 0 0 0 0 0 ⎦⎥
Since the augmented matrix is in reduced
row echelon form, the solution is seen to be
x 1 = -2r
x 2 = s
x 3 = − 3r
x 4 = − 4 r
x 5 = r,
where r and s are any real numbers

Relationship Between Nonhomogeneous Linear

Sytems and Homogeneous Linear Systems
• Let AX=B, B≠0, be a consistent linear system.

If xp is a particular solution to the given nonhomogeneous

system and xh is a solution to the AX=0 homogeneous
system, Then xp+ xh is a solution to the given system
AX=B.

Moreover, every solution X to the nonhomogeneous linear

system AX=B can be written as xp+ xh where xp is a
particular solution to the given nonhomogeneous system
and xh is a solution to the AX=0 homogeneous system.

55
 ELEMENTARY MATRICES,
FINDING A-1
• Remember
• An nxn matrix A is called nonsingular
g ((invertible),
), if there exists an
nxn matrix B such that AB=BA=In, such a B is called an inverse of
A. Otherwise, A is called singular(noninvertible)
• Remember
Definition 1.11: ELEMENTARY ROW(COLUMN) OPERATIONS
An elementary row(column) operation on a matrix A is any one of the
following operations:
(a) Type I: Interchange any two rows (columns)
(b) Type II: Multiply a row (column) by a nonzero number
(c) Type III: Add a multiple of one row (column) to another.

ELEMENTARY MATRIX
• Definition 1.13
• An elementary matrix type I, type II or type
III is a matrix obtained from the identity
matrix In by performing a single
elementary row or column operation of
type I, type II or type III, respectively.

56
 Example
⎡0 0 1 ⎤
⎯ → E 1 = ⎢⎢ 0 1 0 ⎥⎥ Matrix E 1 is of type I
R1↔ R 3
I3 ⎯ ⎯ ⎯
⎢⎣ 1 0 0 ⎥⎦
We interchang e the 1st. and 3rd. rows of I 3
⎡1 0 0⎤
I 3 ⎯ −⎯⎯ → E 2 = ⎢⎢ 0 − 2 0 ⎥⎥ Matrix E 2 is of type II
2R 2

⎢⎣ 0 0 1 ⎥⎦
We multiplied the 2nd row of I 3
⎡1 2 0 ⎤
⎯ → E 3 = ⎢⎢ 0 1 0 ⎥⎥ Matrix E 3 is of type III
2R 2 +R1
I3 ⎯ ⎯ ⎯
R1
⎢⎣ 0 0 1 ⎥⎦
We added twice the 2nd row of I 3 to the 1st row of I 3

• Theorem 1.13
• Let
L t A be
b an mxn matrix t i andd llett an
elementary row(column) operation of type
I, type II, or type III be performed on A to
yield matrix B.
• Let E be the elementary matrix obtained
f
from Im(In) b
by performing
f i th
the same
elementary row (column) operation as was
performed on A. Then B=EA (B=AE)

57
• Example
⎡1 3 1⎤ 2 ⎡− 5 3 0 − 3⎤
Let A = ⎢⎢ - 1 2 3 4 ⎥ ⎯ ⎯ ⎯⎯ → B = ⎢⎢ − 1
⎥ −2R 3+R1
2 3 4 ⎥⎥
⎣⎢ 3 0 1 2 ⎦⎥ ⎢⎣ 3 0 1 2 ⎥⎦
Now
Now, let E be the matrix tha t is obtained from I3
by adding ( − 2) times the 3rd. row of I 3 to the 1st. row
of I 3 . Thus
⎡1 0 - 2⎤
E = ⎢⎢ 0 1 0 ⎥⎥ (I 3 ⎯ −⎯
2R 3+R1
⎯⎯ → E )
⎢⎣ 0 0 1 ⎥⎦
B = EA , lets verify th is
⎡− 5 3 0 − 3⎤ ⎡1 0 − 2⎤ ⎡ 1 3 2 1⎤
⎢−1 2 3 4 ⎥⎥ = ⎢⎢ 0
?
1 0 ⎥⎥ . ⎢⎢ − 1 2 3 4 ⎥⎥
⎢
⎣⎢ 3 0 1 2 ⎦⎥ ⎢⎣ 0 0 1 ⎦⎥ ⎣⎢ 3 0 1 2 ⎦⎥
⎡ − 5 3 0 − 3⎤
= ⎢⎢ − 1 2 3 4 ⎥⎥
⎣⎢ 3 0 1 2 ⎦⎥

• Theorem 1.14
If A and B are mxn matrices, then A is row
equivalent to B if and only if there exist
elementaryy matrices E1, E2,......,Ek such that
B=Ek.Ek-1 .......E2.E1.A

• Proof
If B=Ek.Ek-1 .......E2.E1.A where Ei are elementary
matrices then B results from A by a sequence of
elementary row operations, which implies that A
is row equivalent to B.

58
Theorem 1.15
An elementary matrix E is nonsingular and its
inverse is an elementary matrix of the same
type.
type
⎡0 1 0⎤ ⎡ e 11 e 12 e 13 ⎤
E 1 = ⎢⎢ 1 0 0 ⎥⎥ E -1
1 = ⎢⎢ e 21 e 22 e 23 ⎥⎥
⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ e 31 e 32 e 33 ⎥⎦
⎡1 0 0⎤
E 1 .E −1
1 = I3 = ⎢⎢ 0 1 0 ⎥⎥
⎣⎢ 0 0 1 ⎥⎦
e 21 = 1, e 12 = 1, e 33 = 1
⎡0 1 0⎤
E -1
1 = ⎢⎢ 1 0 0 ⎥⎥
⎢⎣ 0 0 1 ⎥⎦

Lemma 1.1.
L tAb
Let be an nxn matrix
t i and
d llett th
the h
homogeneous
system AX=0 have only the trivial solution X=0.
Then A is row equivalent to In.
Then we can write
In= Ek.Ek-1 .......E2.E1.A
Theorem 1.16
A is nonsingular if and only if A is a product of
elementary matrices.

59
A.B=B.A=In
In= Ek.Ek-1 .......E2.E1.A
A= (Ek.Ek-1 .......E2.E1)-1
= E1-1.E2-1 .......Ek-1-1.Ek-1

Since the inverse of an elementary matrix is an

elementary matrix (Theorem 1.15), we have
established the result

• Corollary 1.3
A is nonsingular if and only if A is row
equivalent to In

• Theorem 1.17
The homogeneous system of n linear
equations in n unknowns AX=0 has a
nontrivial solution if and only if A is
singular.

60
• Example
⎡1 2 ⎤
Let A = ⎢ ⎥ be a singular matrix
⎣ 2 4⎦
Consider the homogeneous system AX = 0;
⎡1 2 ⎤ ⎡ x 1 ⎤ ⎡0 ⎤
That is ⎢ ⎥.⎢ ⎥ = ⎢ ⎥
⎣ 2 4 ⎦ ⎣ x 2 ⎦ ⎣0 ⎦
The reduced row echelon form of the augmented matrix is
⎡1 2 0⎤ − 2 R1 + R 2 ⎡1 2 0⎤
⎢ ⎥ ⎯⎯R⎯⎯→ ⎢ ⎥
⎣ 2 4 0⎦ 2
⎣0 0 0 ⎦
So, a solution is x1 + 2x 2 = 0
x1 = -2x 2
x1 = −2r
x 2 = r , any real number
The homogeneous system has a nontrivial solution.

NOTE THAT
• The following statements are equivalent
f an nxn matrix
for t i A.
A
• 1) A is nonsingular.
• 2) AX=0 has only the trivial solution.
• 3) A is row equivalent to In.
• 4) The linear system AX=B has a unique
solution for every nx1 matrix B.
• 5) A is a product of elementary matrices.

61
 FINDING A-1 by ELEMENTARY
ROW OPERATIONS
We know that

In= Ek.EEk-1 .......E

E2.E
E1.AA
A= (Ek.Ek-1 .......E2.E1)-1
= E1-1.E2-1 .......Ek-1-1.Ek-1
from Theorem 1.16. A is nonsingular and

A-1= (E1-1.E2-1 .......Ek-1-1.Ek-1)-1

= Ek.Ek-1 .......E2.E1

This now provides an algorithm for finding A-1.

Thus, we perform elementary row

operations on A until we get In; the product
of elementary matrices then gives A-1

(Ek.Ek-1 .......E2.E1).[A|In]
=[Ek.Ek-1 .......E2.E1A| Ek.Ek-1 .......E2.E1 In]
=[A-1A| Ek.Ek-1 .......E2.E1]
=[In |A-1]

62
Example
1)Find the A inverse if there
exists.
2) Show that whether A is ⎡1 1 1⎤
A = ⎢⎢0 2 3⎥⎥
singular or not.
3) Show that whether the
linear system AX=B has a
unique solution for every
nx1 matrix B.
⎢⎣5 5 1⎥⎦
4) Whether AX=O has only
trivial solution ?

⎡1 1 1 1 0 0 ⎤ ⎡1 1 1 1 0 0 ⎤
⎢ ⎥ −5 R 1 + R 3 ⎢ ⎥ 1/ 2 ) R 2
⎢0 2 3 0 1 0⎥ ⎯⎯R⎯ ⎯→⎢0 2 3 0 1 0⎥ ⎯(⎯ ⎯→
⎢⎣5 5 1 0 0 1⎥⎦ ⎢⎣0 0 − 4 − 5 0 1⎥⎦
3

⎡1 1 1 1 0 0⎤ ⎡1 1 1 1 0 0 ⎤
⎢ ⎥ ( −1/ 4 ) R 3 ⎢ ⎥ −3 / 2 ) R 3 + R 2
⎢ 0 1 3 / 2 0 1 / 2 0 ⎥ ⎯ ⎯ ⎯ ⎯ → ⎢ 0 1 3 / 2 0 1 / 2 0 ⎥ ⎯(⎯ ⎯ ⎯⎯→ .....
⎢⎣0 0 − 4 − 5 0 1⎦⎥ ⎢⎣0 0 1 5 / 4 0 − 1 / 4⎥⎦
⎡1 1 0 − 1 / 4 0 1/ 4 ⎤
⎢ ⎥
...... ⎯−⎯
R 3 + R1
⎯→ ⎢0 1 0 − 15 / 8 1 / 2 3 / 8 ⎥ ⎯−⎯
⎯ R 2 + R1
⎯
⎯→
⎢⎣0 0 1 5 / 4 0 − 1 / 4⎥⎦
⎡1 0 0 13 / 8 − 1 / 2 − 1 / 8 ⎤
⎢ ⎥
⎢0 1 0 − 15 / 8 1 / 2 3/8 ⎥
⎢⎣0 0 1 5 / 4 0 − 1 / 4⎥⎦
⎡ 13 / 8 − 1 / 2 − 1 / 8⎤
Hence A -1 = ⎢⎢− 15 / 8 1 / 2 3 / 8 ⎥⎥
⎢⎣ 5 / 4 0 − 1 / 4⎥⎦
We can readily verify that A.A -1 = A -1.A = I 3

63
• Theorem 1.18
A nxn matrix
An t i A iis singular
i l if and
d only
l if A iis row
equivalent to a matrix B that has a row of zeros.

Proof
If A is singular, than A is not row equivalent to In.
Thus A is row equivalent to a matrix B≠In which
is in reduced row echeolon form. B must have a
row of zeros

Example
Sh
Show that
th t whether
h th A iis
singular or not.
⎡1 2 − 3⎤
A = ⎢⎢1 − 2 1 ⎥⎥
⎢⎣5 − 2 − 3⎥⎦

64
⎡1 2 − 3 1 0 0⎤ ⎡1 2 − 3 1 0 0⎤
⎢1 − 2 1 0 1 0⎥ ⎯−⎯ ⎯→ ⎢⎢0 − 4
R1+ R 2
⎯ 4 − 1 1 0⎥⎥ ⎯−⎯
5 R1+ R 3
⎯⎯→
⎢ ⎥ R2
⎢⎣5 − 2 − 3 0 0 1⎥⎦ ⎢⎣5 − 2 − 3 0 0 1⎥⎦
⎡1 2 − 3 1 0 0⎤ ⎡1 2 −3 1 0 0⎤
⎢0 − 4 4 − 1 1 0⎥ ⎯−⎯ ⎯⎯→ ⎢⎢0
3 R 2+ R 3
− 4 4 − 1 1 0⎥⎥
⎢ ⎥
⎢⎣0 − 12 12 − 5 0 1⎥⎦ ⎢⎣0 0 0 − 2 − 3 1⎥⎦
At this point A is row equivalent to B
⎡1 2 − 3⎤
B = ⎢0 − 4 4 ⎥
⎢ ⎥
⎣⎢0 0 0 ⎦⎥
Since B has a row of zeros, we stop and
conclude that A is s singular matrix

65