0% found this document useful (0 votes)
65 views

Determinant: Objective: To Introduce The Notion of Determinant, and Some of Its Properties As Well As Applications

This document introduces determinants and their properties and applications. It begins by defining determinants for 1x1, 2x2, and 3x3 matrices. It then generalizes the definition to nxn matrices and introduces minors and cofactors. The document proves several theorems about properties of determinants, including that the determinant of the transpose of a matrix is equal to the determinant of the original matrix. It also proves that the determinant of a product of matrices is equal to the product of the determinants. The document concludes by introducing Cramer's rule for solving systems of linear equations using determinants.

Uploaded by

Mas Gund
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
65 views

Determinant: Objective: To Introduce The Notion of Determinant, and Some of Its Properties As Well As Applications

This document introduces determinants and their properties and applications. It begins by defining determinants for 1x1, 2x2, and 3x3 matrices. It then generalizes the definition to nxn matrices and introduces minors and cofactors. The document proves several theorems about properties of determinants, including that the determinant of the transpose of a matrix is equal to the determinant of the original matrix. It also proves that the determinant of a product of matrices is equal to the product of the determinants. The document concludes by introducing Cramer's rule for solving systems of linear equations using determinants.

Uploaded by

Mas Gund
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 37

Chapter 2

Determinant
 Objective:
To introduce the notion of determinant,
and some of its properties as well as
applications.
Several applications of determinant

 Test for singularity of a matrix instead of by definiti


on.
 Find the area of a parallelogram generated by two vec
tors.
 Find the volume of a parallelopipe spanned by three v
ectors.
 Solve Ax=b by Cramer’s rule.
Introduction to Determinant
(to determine the singularity of a matrix)

11
Consider A  ( a )  F .

If we define det(A)=a,۹ det A 0.


then A is nonsingular.
Case2 2×2 Matrices

 a11 a12  2 2
 Let A    F
a21 a22 
 Suppose a11  0 ,

E12 ( a21 ) E2 (a11 ) a11 a12 


then A 0 a11a22  a12 a21 

If we define det( A)  a11a12  a12a21.


then A is nonsingular  det( A)  0.
Case2 2×2 Matrices (cont.)

 Suppose a11  0 but a12  0 ,

E12 a21 a22 


then A 0 and det( A)  a12a21.
 a12 

Thus A is nonsingular  det( A)  0.


 Suppose a  a  0  A is singular & det(A)=0.
11 21

 To summarize, A is nonsingular  det( A)  0


Case3 3×3 Matrices

 a11 a12 a13 


 Let A  a21 a22 a23   F 33
 

 a31 a32 a33 

-Suppose a11  0,
 a11 a12 a13 
E13 (  aa1131 ) E12 (  aa1121 )  0 a11 a 22  a12 a 21 a11 a 23  a 21a13 
A  a11 a11 
 0 a13 a32  a31a12 a11 a33  a31a13 
 a11 a11 
Case3 3×3 Matrices (cont.)

 From 2x2 case,


 a11a22a11a12a21 a11a23  a21a13

I  det  a13a32 a31a12 0
a11
A a11a33  a31a13
 a11 a11 
define
� a11a22 a33  a11a32 a23  a12 a21a33
+ a12 a31a23 + a13a21a32  a13a31a22  det( A) �0

Then A is nonsingular  det( A)  0.


Easily Shown for Cases

(i )a11  0, a21  0
(ii )a11  a21  0, a31  0
(iii )a11  a21  a31  0

that A I  det( A)  0.
Recall

 For A  F 22 ,

det A  a11a22  a12a21


2
 a11 det( M 11 )  a12 det( M 12 )   ( 1)1+ j a1 j det( M 1 j )
j 1

where M 11  ( a22 ) & M 12  (a21 )


Recall (cont.)

 For A  F 33,

det A  a11 (a22a33  a32a23 )  a12 (a21a33  a31a23 )


+ a13 (a21a32  a31a22 )
 a11 det( M 11 )  a12 det( M 12 ) + a13 det( M 13 )
n
  ( 1)1+ j a1 j det( M 1 j )
j 1

a22 a23  a21 a23  a21 a22 


where M 11   , M 12   , M 13  

a32 a33  a31 a33  a31 a32 
Generalization

, n > 3 , and let


n n
Definition: Let A  F

 M ij  F ( n 1)( n 1) the matrix obtained from A by


deleting the row & column containing aij .
 The det( M ij ) is called the minor of aij
 The cofactorAij ofaij is denoted as
Aij  ( 1)i + j det( M ij )
n n
Definition: The determinant of A  F is defined as

a11 , if n=1
det A  n

a
j 1
1j A1 j
, if n>1

n n
Note: det is a function from F to F .
Theroem2.1.1:Let A  F nn , n  2 ,
n n
det( A)  �aik Aik  �akj Akj
k 1 k 1

for i  1, 2,...n. and j  1, 2,...n.

Hint: By induction or sign-type definition.


Theroem2.1.2: Let A  F
n n
,and det( AT
)  det( A)
Pf: By induction,
n=1,ok!
Suppose the theorem is true for n=k.
If n=k+1, k +1 k +1
det( A)   ( 1) j +1 a1 j det( M 1 j )   ( 1) j +1 a1 j det( M 1Tj )
j 1 j 1

 det( AT ) By induction

The result then follows.


n n
Theroem2.1.3: Let A  F n
be a triangular matrix.
Then det( A)   aii .
i 1
Hint:expansion for lst row or column and induction on n
.

Theroem2.1.4: (i)If A has a row or column det(consisting


A)  0.
entirely of zeros, then
(ii)If A has two
det( A) identical
 0. rows or columns,
then
§2-2 Properties of Determinants

Note that det( A + B )  det A + det B

For example, A  1 0, B  0 0


 0 0 0 1 
   

Question: Is
det( AB )  det( A) det( B ) ?
Lemma2.2.1:
Let A  F nn , then
n

a
k 1
ik Ajk  ai1 Aj1 +  + ain Ajn

det( A) , if i  j

0 , if i  j
Proof of Lemma2.2.1

Pf: Case for i=j follows directly from the definition


of determinant.
For i  j ,

define A to be the matrix obtained
from A by replacing the jth row of A by ith row

of A. (Then A has two identical rows)
 0  det( A )  ai1 Aj1 +  + ain Ajn
expansion along jth row

 ai1 Aj1 +  + ain Ajn


Proof of Lemma2.2.1(cont.)

�a11 a12 K a1n �


� �
�M M M�
�ai1 ai 2 K ain �
 � �
A  �M M M�
�ai1 ai 2 K ain �jth row
� �
�M M M�
�an1 an 2 K ann �
� �
Note that
 det( Ei ( ))    det( I )  by Th. 2.1.3

 det( Eij ( ))  1  by Th. 2.1.3

 det( Eij )  1  先對非交換列展開


數學歸納法
n
 det( Ei ( ) A)  a
j 1
ij Aij   det( A)

 det( Ei ( )) det( A)

 det( Eij ( ) A)   (a


k 1
ik +a jk ) A jk

 a ik A jk +  a jk Ajk
0
Lemma 2.2.1
 det( A)  det( Eij ( )) det( A)
  
 a ( j , :)

  ith row
det( Eij A)  det   
 
 a ( i , 
:)  jth row

   

  
  a (i , :)
E ji ( 1) Eij (1) E ji ( 1)
  ith row
 det      det( A)  det( Eij ) det( A)
 
 a ( j , :) jth row

  

Thus, we have

det( EA)  det( E ) det( A)

If E is an elementary matrix
In fact, det(AE)=det(A)det(E)

Question: det( AB )  det( A) det( B ) ?


Theorem2.2.2: A  F nn is singular  det( A)  0.
Pf:Transform A to its row echelor from as
U  Ek  E1 A
k
� det(U )  (�det( Ei )) det( A)
i 1 0
If A is singular
 U contains a zero row.
 det(U )  0  det( A)  0
If A is nonsingular
 all the diagonal elements of U is 1.
 det(U )  1
 det( A)  0
The result then follows.
n n
Theorem2.2.3:Let A, B  F .Then

det( AB )  det( A) det( B )


Pf: If B is singular
 AB is singular
Th.2.2.2
 det( AB )  0  det( A) det( B )
If B is nonsingular
 B  Ek  E1 , where Ei are elementary matrices.
 det( AB )  det( AEk  E1 )

 det( A) det( Ei )

 det( A) det( Ei )

 det( A) det( B )
§2-3 Cramer’s Rule

 Objective:
1
Use determinant to compute A and
solve Ax=b.
The Adjoint of a Matrix

n n
Def: Let A  F .The adjoint of A is defined to be
�A11 A21 L An1 �
�A A �
An 2 �
adj ( A)  � 12 22

�M O M�
� �
�A1n A2 n L Ann �
where Aij are cofactor of aij .
By Lemma2.2.1, we have
n
det( A) if i  j
 aik Ajk 
k 1
0 if i  j

� A(adj ( A))  det( A) �


I

If A is nonsingular, det(A) is a nonzero scalar

� A( det(1 A) adj ( A))  I


1
�A  1
det( A ) adj ( A)
Example 1 (P.116)

�a22 a12 �
For a 2×2 matrix : adj ( A)  �
�a21 a11 �

If A is nonsingular ޹ det( A) 0 , then

1 �a a 12 �
A  det(1 A) � 22 �
a21
� a11 �
Example 2 (P.116)

2 1 2
Q: Let A   3 2 2 , compute adj A and A-1.

1 2 3
Sol: �2 1 2 �
adj ( A)  �
�7 4 2��

�4 3 1 � �
�2 1 2 �
A1  det(1 A) adj ( A)  15 �
�7 4 2 �

�4 3 1 �
� �
Theorem2.3.1:(Cramer’s Rule) v
n n
Let A  F be nonsingular and b ޹ F . Denote Ai the
n
v
matrix obtained by replacing the ith column of A by b.Then
v v
the unique sol. of Ax  b is
det( Ai )
xi  det( A ) , for i  1,  , n
v 1
v v
Pf: Qx A b  1
det( A ) adj ( A) �
b
n
� xi  1
det( A ) �b
j 1
j < adjA >ij

n
det( Ai )
 1
det( A ) �b j A ji  det( A)
j 1
Example 3 (P.117)

Q: Use Cramer’s rule to Solve

x1 + 2 x2 + x3  5
2 x1 + 2 x2 + x3  6
x1 + 2 x2 + 3 x3  9
Example 3 (cont.)

Sol:
det( A)  4
5 2 1
 
det( A1 )  det 6 2 1  4
 
9 2 3
det( A2 )  4
det( A3 )  8
 x1  det( A1 )
det( A )  1, x2  det( A2 )
det( A )  1, x3  det( A3 )
det( A ) 2
  
 Let a , b , c  3 .Then volume of the parallelopipe
  
spanned by a , b and c is
   
V  a  (b  c )  det(a b c )
 
 Let a , b   2
.Then the area of the parallelogram
 
spanned by a and b is

A  det( ab )
Application 1: Coded Message (P.118)

For example, the message


Send Money
might be coded as
5, 8, 10, 21, 7, 2, 10, 8, 3

here the S is represented by a “5”, the E is represented


by a “8”, and so on.
Application 1: Coded Message (cont.)

If A is a matrix whose entries are all integers and


whose determinants is ± 1, then, since A1  �adj A ,
the entries of A-1 will be integers.

Let
1 2 1�
� �5 21 10 �
� � � �
A�2 5 3 � and B= �8 7 8 �
�2 3 2 � �10 2 3 �
� � � �
Application 1: Coded Message (cont.)

We can decode it by multiplying by A-1

�1 1 1 � �31 37 29 � �5 21 10 �
� �
� �� �
� 2 0 1 �
�80 83 69 
�� 8 7 8 �
� 4 1 1 �
�54 67 50 �� 10 2 3 �
� �
� �� �
We can construct A
-1by applying a sequence of row operations on identity matrix.
A AB(encoding Message)
Note:

det( A)  �det( I )  �1

You might also like