Transcript Slide 1
Recap
Row and Reduced Row Echelon
Elementary Matrices
If m and n are positive integers, then
an m n matrix is a rectangular
array in which each entry aij of the
matrix is a number. The matrix has m
rows and n columns.
a1,1
a2,1
a3,1
am ,1
a1,2
a1,3
a2, 2
a3,2
a2,3
a3,3
am,2
am,3
a1,n
a2,n
a3, n
am ,n
A real matrix is a matrix all of whose
entries are real numbers.
i (j) is called the row (column)
subscript.
An mn matrix is said to be of size (or
dimension) mn.
If m=n the matrix is square of order n.
The ai,i’s are the diagonal entries.
Given a system of equations we can
talk about its coefficient matrix and its
augmented matrix.
These are really just shorthand ways of
expressing the information in the
system.
To solve the system we can now use
row operations instead of equation
operations to put the augmented
matrix in row echelon form.
1. Interchange two rows.
2. Multiply a row by a nonzero constant.
3. Add a multiple of a row to another row.
Two matrices are said to be row
equivalent if one can be obtained
from the other using elementary row
operations.
A matrix is in row-echelon form if:
› All rows consisting entirely of zeros are at
the bottom.
› In each row that is not all zeros the first
entry is a 1.
› In two successive nonzero rows, the
leading 1 in the higher row is further left
than the leading 1 in the lower row.
1. Write the augmented matrix of the
system.
2. Use elementary row operations to find a
row equivalent matrix in row-echelon
form.
3. Write the system of equations
corresponding to the matrix in rowechelon form.
4. Use back-substitution to find the solutions
to this system.
In Gauss-Jordan elimination, we
continue the reduction of the
augmented matrix until we get a row
equivalent matrix in reduced rowechelon form. (r-e form where every
column with a leading 1 has rest zeros)
1
0
0
0 0
1 0
0 1
a
b
c
A system of linear equations in which
all of the constant terms is zero is
called homogeneous.
All homogeneous systems have the
solutions where all variables are set to
zero. This is called the trivial solution.
New Stuff
Using Elementary Matrices
An n by n matrix is called an elementary
matrix if it can be obtained from In by a
single elementary row operation.
These matrices allow us to do row
operations with matrix multiplication.
Theorem: Let E be the elementary matrix
obtained by performing an elementary
row operation on In. If that same row
operation is performed on an m by n
matrix A, then the resulting matrix is given
by the product EA.
Three types of Elementary
Matrices
These correspond to the three types of
EROs that we can do:
› Interchanging rows of I -> Type I EM
› Multiplying a row of I by a constant -> Type II
EM
› Adding a multiple of one row to another ->
Type III EM
Type I EM
0 1 0
1
0
0
0 0 1
E1 =
How is this created?
Eg. 1 Suppose
A=
a11
a21
a
31
a12
a22
a32
E1A =
0 1 0 a11
1
0
0
a21
0 0 1 a31
a22
a32
=
a21
a11
a
31
a13
a23
a33
a12
a22
a12
a32
a23
a13
a33
What is AE1?
a13
a23
a33
Type II EM
E2 =
E2A =
AE2 =
1 0 0
0 1 0
0 0 3
1 0 0 a11
a
0 1 0 21
0 0 3 a31
a22
a32
a11
a21
a
31
1 0 0
0 1 0
0 0 3
a12
a22
a32
a22
a23
a33
a12
a22
a23
a33
=
=
a12
a11
a21 a22
3a
31 3a32
a11
a21
a
31
a12
a 22
a32
a22
a23
3a33
3a13
3a23
3a33
Type III EM
E3 =
E3A =
AE3 =
1 0 3
0
1
0
0 0 1
1 0 3 a11 a12 a22
0
1
0
a21 a22 a23
0 0 1 a
31 a32 a33
a11
a21
a
31
a12
a22
a32
a22
a23
a33
1 0 3
0
1
0
0 0 1
=
a11 3a31
a21
a
31
a12 3a32
= a11 a12
a21 a22
a
31 a32
a22
a32
a13 3a33
a23
a33
3a11 a13
3a21 a23
3a31 a33
Let A and B be m by n matrices. Matrix B
is row equivalent to A if there exists a
finite number of elementary matrices E1,
E2, ... Ek such that
B = EkEk-1 . . . E2E1A.
Break it down
This means that B is row equivalent to A if
B can be obtained from A through a
series of finite row operations.
If we then take two augmented matrices
(A|b) and (B|c) and they are row
equivalent, then Ax = b and Bx=c must
be equivalent series
Break it down
If A is row equivalent to B, B is row
equivalent to A
If A is row equivalent to B and B is row
equivalent to C then A is row equivalent
to C
Example
Compute the inverse of A for A =
4 3 1 0 0
1
1
2
0
|
0
1
0
2
2 3 0 0 1
3 1 0 0
1 4
0
2
3
|
1
1
0
0 6 3 2 0 1
3
1
1
1 4 3 1 0 0 1 4 0 2
2
2
0 2 3 |1 1 0 0 2 0 | 12 12 12
0 0 6 1 3 1 0 0 6 1
3
1
4 3
1
1 2 0
2
2 3
Example
1
1 1 0 0 1
1
1
1 0 0 1
2
2
2
2
2
2
1
1 0 1 0 | 1
1
1
0 2 0 | 1
2
2
2
4
4
4
1
1
3
1 0 0 1 1
0 0 6 1
6
2
6
Now, solve the system:
x1 4 x2 3x3 12
x1 2 x2 12
2 x1 2 x2 3x3 8
Example
We can employ the format Ax = b so x=A-1b
We just calculated A-1 and b is the column
vector 12
12
8
So we can easily find the values of x by
multiplying the two matrices
Determinants
Keys to calculating Inverses
Require square matrices
Each square matrix has a determinant
written as det(A) or |A|
Determinants will be used to:
› characterize on-singular matrices
› express solutions to non-singular systems
› calculate dimension of subspaces
If A and B are square
then AB A B
It is not difficult to
appreciate that
AT A
If A has a row (or
column) of zeros
then
A 0
If A has two identical
rows (or columns)
then
1 7 1
2
0
2 0
3
5
3
C1 C2
If B is obtained from A by ERO,
Ri R j (Ci C j ) interchanging two
rows (or columns) then
B A
If B is obtained from A by ERO where row
(or column) of A were multiplied by a
scalar k, then B k A
If B is obtained from A by ERO where a
multiple of a row (or column) of A were
added to another row (or column) of A
then A B
A
a11
a12
a21
a22
a11 a22 a12 a21
That Is the determinant is equal to the
product of the elements along the
diagonal minus the product of the
elements along the off-diagonal.
3 2
A
1 5
A (3)(2) (1)(6) 0
Note: The matrix A is said to be
invertible or non-singular if det(A)≠ 0. If
det(A) = 0, then A is singular.
For a 3 by 3 matrix
Using EROs on rows 2 and 3
a11
a21
a
31
a12
a22
a32
a13
a23
a33
a11
0
0
a12
a11a22 a21a12
a11
a11a32 a31a12
a11
a13
a11a23 a21a13
a11
a11a33 a31a13
a11
For a 3 by 3 matrix
The matrix will be row equivalent to I iff:
a11a22 a21a12
a11
a11
a11a32 a31a12
a11
a11a23 a21a13
a11
0
a11a33 a31a13
a11
For a 3 by 3 matrix
This implies that the
Det(A) =
a11a22a33 a11a32a23 a12a21a33 a12a31a23 a13a21a32 a13a31a22
Use EROs to find:
3
2
1
5
5
4
0
2
6
STEP 1: Apply C1 C3
from property 5 this
gives us B A
1 2 3
2 4 5
6 0
5
STEP 2: Convert
matrix to Echelon
form
1
R2 R2 2R1
R3 R3 6R1
2
3
0
0
1
0 12 23
1
2
3
R2 R3 0 12 23
0
0
1
Therefore
the same
as:
1
2
3 2 1is
5 4 2
5 0 6
3
0 12 23 1(12)(1) 12
0 0
1
matrix is now in echelon
form so we can multiply
elements of main
diagonal to get
determinant
1 x
x2
Factorize the determinants of 1 y
2
1 z
1 2 4
What is
1 1 1
?
1 3 9
y
z2
We see that y – x is a factor of row 2 and
z – x is a factor of row 3 so we factor
them out from:
x
x2
yx
y2 x2
zx
z2 x2
1
R2 R2 R1 0
R3 R3 R1 0
And we get:
( y x)(z x)
1 x
x2
0 1
yx
0 1
zx
R3 R3 R2
( y x)(z x)
1 x
x2
0 1
yx
0 0
zy
The matrix is now in echelon form so we
can multiply elements of main diagonal
to get determinant and then multiply by
factors to get:
1 x
x2
1 y
y2
1 z
z2
=
( y x)(z x)(z y)
1 2 4
Now, the matrix 1
1 1 corresponds to
1 3 9
x 2
y 1
z 3
Since
1 x
x2
1 y
y2
1 z
z2
= ( y x)(z x)(z y)
Then
1 2 4
1 1 1
1 3 9
= (1 (2))(3 (2))(3 1) 3(1)(4) 12
Cofactor expansion is one method used
to find the determinant of matrices of
order higher than 2.
If A is a square matrix, then the minor Mi,j
of the element ai,j of A is the determinant
of the matrix obtained by deleting the ith
row and the jth column from A.
Consider the matrix
3 2 1
A 5 4 2 .
5 0 6
The minor of the entry “0” is found by
deleting the row and the column
associated with the entry “0”.
The minor of the entry “0” is
3 1
5 2
3(2) 1(5) 1
Note: Since the 3 x 3 matrix A has 9
elements there would be 9 minors
associated with the matrix.
The cofactor Ci,j = (-1)i+jMi,j.
(1) i j
1, i j even
1, i j odd
Since
we can think of the
cofactor of as nothing more than its signed
minor.
Find the minor and cofactor of the entry
3 2 1
“2” for
A 5 4 2
5 0 6
We first need to delete the row and
column corresponding to the entry “2”
The Minor of 2 is
5 2
5(6) 5(2) 40
5 6
The minor corresponds to row 1 and
column 2 so applying the formula, we
have cij (1)12 40 1(40) 40
So the cofactor of the entry “2” is 40.
Theorem: Let A be a square matrix of
order n. Then for any i,j,
n
Columns:
det (A) A ai , jCi, j
j 1
and
n
Rows: det (A) A ai , jCi, j .
i1
3 2 1
A 5 4 2
5 0 6
Given
Cofactor is found for the first entry in
column 1 “-3”
4 2
c11
find det(A).
0 6
24 0 24
Cofactor is found for the second entry in
column 1 “-5” c 2 1 12 0 12
21
0
6
Cofactor is found for the third entry in
column 1 “5”
c31
2
4
1
44 0
2
The cofactors are then multiplied by the
corresponding entry and summed.
A 3(c11 ) 5c21 5c31
3( 2 4) 5(1 2) 5(0)
7 2 6 0
1 2
Using row 2 - expansion we fix row 2 and
find the minors for each entry in row 2
then apply the sign corresponding to
each entries position to find the
cofactors. The cofactors are then
multiplied by the corresponding entry
and summed.
A 5(c22 ) 4c22 2c23
5 *
2 1
0 6
4*
3 1
5
6
2*
3 2
5
5 * (12) 4 * (23) 2 * (10)
60 92 20
12
0
It is easy to show that
a11
a12
a13
a14
0
0
a 22
0
a 23
a33
a 24
a11 a 22 a33 a 44
a34
0
0
0
a 44
If A is square and is in Echelon form then
is the product of the entries on the
(main) diagonal.
1
0
0
0
0 0
2 0
0 1
0 0
2
0
1* 2 * 1* 0 0
0
0
Using CRAMER’S RULE we can apply this
method to finding the solution to a
system of linear systems that have the
same number of variables as equations
There are two cases to consider
Consider the square system AX
= B where A is n x n.
If A 0 then the system has
either I) No solution or ii) Many
solutions
If A1 is formed from A by
replacing column 1 of A with
column B and
I. A1 0 , then the system has
NO solution
II. |A| ≠ 0, then the system has
a unique solution
Consider the square system AX
= B where A is n x n.
If |A|≠ 0 then the system has
a unique solution
The unique solution is obtained
by using the Cramer’s rule.
Where Ai is found from A by replacing
column i of A with B.
X1
A1
X2
A2
X3
A3
A
A
A
etc
Use Cramer’s rule to write down the
solution to the system
5 x1 3x2 7
4 x1 5 x2 3
5 x1 3x2 7
4 x1 5 x2 3
A 2512 37
A 2512 37
A1
A2
7
3
3
5
5
7
4 3
26
43
X1
A1
A
26
37
43
X2
A
37
26 43
,
37
27
A2
T
adjA
(
C
)
The adjoint of A (adj.A)) isij
: , the transpose of the matrix of
cofactors (a matrix of signed minors).
If A 0 then the inverse of A exists and
A 1
1
adjA
A
If A 0, A has no inverse.
Find inverse of
Det(A) =
3 2 3
A 1 2 1
1 3 0
a11a22a33 a11a32a23 a12a21a33 a12a31a23 a13a21a32 a13a31a22
= (3*2*0) – (3*3*1) – (-2*1*0)+(-2*1*1)
+(3*1*3) –(3*1*2)
=0 – 9 – 0 – 2 + 9 – 6
= -8
adjA (Cij )T
2
3
2
3
2
2
1
0
3
0
3
1
1 1
1
3
1
3
1
0
3
0
3
1
1 2
1 3
3 2
1 3
3 2
1 2
T
0
3 1
3 9 8
9 3 11 1 3 0
8 0
0 11 8
8
T
1
A adjA
A
1
8
3 9
1
1
A 1 3 0
8
0
11
1
a.
b.
c.
Given that
1 4 2
A 1 1 1
2 4 5
show:
A2 4 A 3I 0
0
A
A 13A 12I
3
d.
e. Solve
A 1
1
( AI A)
3
T
T
x
(
x
,
x
,
x
)
b
(
1
,
1
,
1
)
1 2 3 and
A x b where
1 4 2 1 4 2 1 16 8
2
A 1 1 1 1 1 1 4 1 4
2 4 5 2 4 5 8 16 17
1 16 8 1 4 2 1 0 0 0 0 0
2
A 4 A 3I 4 1 4 4 1 1 1 3 0 1 0 0 0 0 0
8 16 17 2 4 5 0 0 1 0 0 0
A 2 4 A 3I 0
A 2 4 A 3I
A( A 4 I ) 3I
A( A 4 I ) 3I
Taking the determinant
Implying that and therefore; exists
A A 4 I 3I
Since, A 4 A 3I 0
Multiplying through by A
2
A3 4 A 2 3 A 0
A3 4 A 2 3 A
( A 2 4 A 3I )
A3 4(4 A 3I ) 3 A
A3 16A 12I 3 A
A3 13A 12I
Multiplying through by A-1
A 2 4 A 3I 0
( AA) A 1 4 AA1 3 A 1 0
( AA1 0)
A 4 I 3 A 1 0
3 A 1 4 I A
A 1
1
( 4 I A)
3
1
A (4 I A)
3
4 0 0 1 4 2
3 4 2
1
1
1
A 0 4 0 1 1 1 1 3
1
3
3
0
0
4
2
4
5
2
4
1
1
Since A x b
1
xA b
3 4 2 1 3 1
1
1
x 1 3 1 1 3 1
3
3
2 4 11 3 1