Beruflich Dokumente
Kultur Dokumente
• 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:
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)
Example:
2
• Linear System (Linear Equations Sytem):
A linear system is a set of m linear equations each in n unknowns.
y
• If the linear system (2)
( ) has no solution, it is said to be INCONSISTENT.
• 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):
We say that (2) and (3) are equivalent if they both have exactly the same
solutions.
4
• Example (Homogeneous Sytems)
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):
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?
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
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:
MATRICES and
MATRIX OPERATIONS
Q. Why we are interested in matrices?
A Because
A. B we can express LES b by using
i ththem.
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 ⎥⎦
• 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⎥⎦
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⎦
⎡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
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
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 ⎦⎥
• Example
⎡− 2 5 ⎤
⎡1 2 − 1⎤
A = ⎢ B = ⎢⎢ 4 − 3 ⎥⎥
⎣3 1 4 ⎥⎦
⎢⎣ 2 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⎦
⎡ 1 2⎤
⎡− 2 3 4⎤
A = ⎢⎢ 3 4⎥⎥ B= ⎢ ⎥
⎣ 3 2 1⎦
⎣⎢−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
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 ⎦⎥
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 ⎦
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⎦
⎡ 3 0 4 − 1⎤
⎢2 1 1 0⎥
⎢ ⎥
⎢⎣− 1 2 3 1 ⎥⎦
⎡ 3 0 4 −1 2 ⎤
⎢ ⎥
⎢ 2 1 1 0 − 4⎥
⎢⎣− 1 2 3 1 5 ⎥⎦
⎡ 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 ⎦
⎡ 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.
• Example
⎡ 1 − 3⎤
⎡ 1 2 −1⎤
If A=⎢ ⎥ then A = ⎢⎢ 2 2 ⎥⎥
T
⎣− 3 2 7 ⎦ ⎢⎣−1 7 ⎥⎦
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⎦
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 ⎥⎦
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 ⎥⎦
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
⎡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
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 ⎦
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 ⎥⎦
30
⎡ B11 B12 ⎤
⎡A
 A12
 A13 ⎤
 B = ⎢⎢B21 B 22 ⎥⎥
A = ⎢ 11 ⎥
Â
⎣ 21 Â 22 Â 23 ⎦ ⎣⎢ B31 B32 ⎥⎦
• 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.
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 ⎦
Theorem 1.5
The inverse of a matrix, if it exists, is unique.
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.)
⎡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
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 ⎥⎦
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 ⎦
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.
• 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 4⎤ 2R ⎡2 4 8⎤
⎢ ⎥
C = ⎢1 - 2 2⎥ → D = ⎢⎢1 - 2 2⎥⎥
1
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⎦
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 ⎦
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.
43
SOLVING LINEAR SYSTEMS
• The method where [C|D] is in row echelon
f
form is
i called
ll d GAUSSIAN ELIMINATION
ELIMINATION.
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 ⎥⎦
⎣
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:
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
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.
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
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
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
• 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
(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