MA11010

FURTHER ALGEBRA AND CALCULUS

Dave Binding
Room: W11
Tel: 2775
E-mail: dmb

For access to module notes and

information:

http://users.aber.ac.uk/dmb/

1
 COURSE CONTENT
Background
Information sources.

Matrix Algebra
Matrix operations (addition, scalar multiplication, matrix
multiplication, transposition, inversion). Special types of
matrices (zero, identity, diagonal, triangular, symmetric, skew-
symmetric, orthogonal). Row equivalence.

Linear equations
Systems of linear equations. Coefficient matrix, augmented
matrix. Elementary row operations. Gaussian and Gauss-Jordan
elimination.

Determinants
Properties of determinants. Computation of determinants.

2
Return to index
 COURSE CONTENT
Partial derivatives
Functions of several variables. Partial Derivatives.
Differentiability and linearisation. The chain rule. Critical points.
Change of variables - the Jacobian.

Multiple integrals
Riemann sums and definite integrals. Double integrals in
rectangular coordinates; areas. Substitution in multiple
integrals.

3
Return to index
LITERATURE AND OTHER INFORMATION
SOURCES

The Physical Sciences

Library

“Just the Maths”

Course notes by A J Hobson
(Ex University of Coventry)
http://www.mis.coventry.ac.uk/jt
m

‘mathcentre’ - Self-study
resources targetted at
individual courses
http://www.mathcentre.ac.uk

4
Return to index
 CALCULATORS

Calculators are NOT permitted

5
Return to index
 MATRICES
An m x n matrix A is an array of elements made up with m
rows and n columns:
 a11 a12 a13 . . . . a1n 
a . . . . a2n 
 21 a22 a23
 a31 a32 a33 . . . . a3 n 
 
A . . . . . . . . 
 . . . . . . . . 
 
 . . . . . . . . 
a . . . . amn 
 m1 am 2 am 3
aij is the element in i-th row and j-th column. That is, the
first index refers to the row number and the second index
refers to the column number.

6
Return to index
 MATRICES
m x n is referred to as the shape of the matrix.
Short-hand notation: A = [aij]
If the entries belong to a field of scalars K, then A is said
to be defined over the field K.
The main diagonal consists of the elements aii.
If m = n the matrix is square.
If n = 1 the matrix is referred to as an m-vector.

Special matrices:
Zero matrix,
All-one matrix,
Identity matrix.

7
Return to index
 ZERO MATRIX
The zero m x n matrix is an array of elements made up
with m rows and n columns with all elements being equal
to zero:

0 0 0 . . . . 0
0 0 0 . . . . 0

0 0 0 . . . . 0
 
0  . . . . . . . .
. . . . . . . .
 
. . . . . . . .
0 0 0 . . . . 0


8
Return to index
 ALL-ONE MATRIX
The all-one m x n matrix is an array of elements made up
with m rows and n columns with all elements being equal
to 1:

1 1 1 . . . . 1
1 1 1 . . . . 1

1 1 1 . . . . 1
 
. . . . . . . .
. . . . . . . .
 
. . . . . . . .
1 1 1 . . . . 1


9
Return to index
 ALL-ONE MATRIX
The identity n x n matrix is a square array of elements
made up with n rows and n columns with all elements on
the principal diagonal being equal to 1 and all other
elements being equal to zero:

1 0 0 . . . 0
0 1 0 . . . . 

0 0 1 . . . .
 
In   . . . . . . .
. . . . . . .
 
. . . . . . .
0 0 0 . . . 1


10
Return to index
 MATRIX OPERATIONS

Equality:

If A is an m x n matrix and B is an p x q matrix, then A = B

if and only if m = p, n = q and aij = bij for all i,j.

That is, for two matrices to be equal, their shapes must

be the same and the corresponding elements must all be
equal.

11
Return to index
 MATRIX OPERATIONS

Scalar multiplication:
If A is an m x n matrix and if  is a scalar then A = [aij]

i.e:  a11 a12 a13 . . . . a1n 

 a . . . . a2n 
 21 a22 a23
 a31 a32 a33 . . . . a3 n 
 
A   . . . . . . . . 
 . . . . . . . . 
 
 . . . . . . . . 
a . . . . amn 
 m1 am 2 am 3

12
Return to index
 MATRIX OPERATIONS

Sum:
If A and B are m x n matrices (they must have the same
‘shape’), then A + B = [aij + bij]

 a11  b11 a12  b12 a13  b13 . . . . a1n  b1n 

 a b a2n  b 2n 
 21 21 a22  b 22 a23  b 23 . . . .
 a31  b 31 a32  b 32 a33  b 33 . . . . a3 n  b 3 n 
 
AB   . . . . . . . . 
 . . . . . . . . 
 
 . . . . . . . . 
a  b amn  b mn 
 m1 m1 am 2  b m 2 am 3  b m 3 . . . .

13
Return to index
 MATRIX OPERATIONS

Difference:
If A and B are m x n matrices (they must have the same
‘shape’), then A  B = [aij  bij]

 a11  b11 a12  b12 a13  b13 . . . . a1n  b1n 

 a b a2n  b 2n 
 21 21 a22  b 22 a23  b 23 . . . .
 a31  b 31 a32  b 32 a33  b 33 . . . . a3 n  b 3 n 
 
A B   . . . . . . . . 
 . . . . . . . . 
 
 . . . . . . . . 
a  b amn  b mn 
 m1 m1 am 2  b m 2 am 3  b m 3 . . . .

14
Return to index
 MATRIX OPERATIONS

Transpose: If A is an m x n matrix, the transpose of A is

an n x m matrix formed by interchanging the rows and
columns, i.e: [AT]ij = [aji]

i.e: a11 a21 a31 . . . . am1 

a a22 a32 . . . . am 2 
 12
a13 a23 a33 . . . . am 3 
 
AT   . . . . . . . . 
 . . . . . . . . 
 
 . . . . . . . . 
a a2n a3 n . . . . amn 
 1n

15
Return to index
 MATRIX OPERATIONS

Product: If A is an m x n matrix and B is a n x p matrix,

then the product AB exists and is an m x p matrix defined
by:
n

 AB   aik bkj 
 k 1 

Note: The no. of columns of A must be the same as the

no. of rows of B

Often, a summation convention is used: When an index is

repeated in a given term it indicates a sum over the
relevant range of integers. Thus the summation sign can
be omitted:
 AB ij  aik bkj
16
Return to index
 MATRIX OPERATIONS

Product: If A is an m x n matrix and B is a n x p matrix,

then the product AB exists and is an m x p matrix defined
by: n
 
 AB   aik bkj 
 k 1 

Written out more fully, the i,j element in the product is:

 AB ij  ai 1b1j  ai 2b2 j ai 3b3 j .....  ain bnj

17
Return to index
 MATRIX OPERATIONS
 1 1 0 2 1 

Example: Let A  2 0 1 and
 0  1
  B  
3  1 2  1 2 

Then AB is a 3 x 2 matrix:
. .
AB  . .
. .

18
Return to index
 MATRIX OPERATIONS
 1 1 0 2 1 

Example: Let A  2 0 1 and
 0  1
  B  
3  1 2  1 2 

Then AB is a 3 x 2 matrix:
. . 
AB  . . 
. 8
The entry (AB)32 is calculated by taking the scalar
product of the 3rd row vector in A with the 2nd column
vector in B:
i.e: (AB)32 = (3 x 1) + (-1 x -1) + (2 x 2) = 8.

19
Return to index
 MATRIX OPERATIONS
 1 1 0 2 1 

Example: Let A  2 0 1 and
 0  1
  B  
3  1 2  1 2 

Then AB is a 3 x 2 matrix:
2 0 
AB  5 4
8 8 
Repeat the process for all elements.

20
Return to index
 RULES OF MATRIX ALGEBRA

If A, B and C are matrices and  and  are scalars, then:

     A  A  A    A    A 
  A  B   A  B   AB   A  B  A B

1A  A,  1A   A

AB B A Commutative law.

 A  B  C  A   B  C Associative law for addition.

A  0  A, A    A   0

21
Return to index
 RULES OF MATRIX ALGEBRA

If A, B and C are matrices and  and  are scalars, then:

 AB C  A BC Associative law for product

 A  B C  AC  BC
Distributive laws
A B  C   AB  AC

AΙ  ΙA  A
A 
T T
 A,  A  T   AT

 A  B T  AT  BT

 AB T  BT AT
22
Return to index
 SPECIAL TYPES OF MATRICES
Assume A is a square matrix:

A is symmetric if aij  a ji ,  i , j

A is anti-symmetric (skew-symmetric) if aij  a ji ,  i , j

A is upper triangular if aij  0,  i  j

A is lower triangular if aij  0,  i  j

A is diagonal if aij  0,  i  j

A is orthogonal if AT A  AAT  I

23
Return to index
 EXAMPLES
 1 2  1  1 1 0
A  2  3 4  B  2 0 1
0  4 5  3  1 2

 1 2  1  1 1 0  2 3  1
A  B  2  3 4   2 0 1   4  3 5 
0  4 5  3  1 2 3  5 7 

 1 2  1  1 1 0  0 1  1
A  B  2  3 4   2 0 1   0  3 3 
0  4 5  3  1 2   3  3 3 

24
Return to index
 EXAMPLES
 1 2  1 1 2 0
A  2 3 4  AT   2 3  4
0  4 5    1 4 5 
Notice that:
2 4  1
A  AT   4 6 0  (Symmetric)
  1 0 10 
0 0  1
A  AT  0 0 8  (Anti-symmetric)
 1  8 0 

25
Return to index
 EXAMPLES
 1 1 0 1
 1 2 3
A  B  2 0 1 C    1
  1 0 1 3  1 2  2 

 1 1 0
 1 2 3   14  2 8
AB     2 0 1   
  1 0 1  2  2 2 
 3  1 2 

(The product BA does not exist)

 1 1 0  1  0 
BC  2 0 1   1   4
3  1 2  2  8 

26
Return to index
 EXAMPLES

1 0 2   1 0 0  1 0 0
A  0 2  4 B  2 0 0 C  0 0 0
0 0  1 3  1 2 0 0 2

A is an upper triangular matrix and B is a lower triangular

matrix. C is a diagonal matrix.

27
Return to index
 EXAMPLES
 1 0 2  1 1 0
A    3 2  4 B  2 0 1
 3  1 0  3  1 2

 1 0 2   1 1 0  7  1 4 
AB    3 2  4 2 0 1    11 1  6
 3  1 0  3  1 2  1 3  1
 1 2 3   1  3 3   7  11 1 
BT AT   1 0  1 0 2  1    1 1 3 
0 1 2  2  4 0   4  6  1

This verifies  AB T  BT AT for this particular example.

28
Return to index
 INVERTIBLE MATRICES
Let A be an n x n matrix. If a matrix A-1 exists, such that
AA-1 = A-1A = I then A is said to be invertible (non-
singular) and A-1 is the inverse of A. If the inverse does
not exist then A is said to be singular.

An invertible matrix has a unique inverse.

Proof: Let B and C be inverses of A. Then:
AB = BA = I, AC = CA = I
Hence: B = BI = B(AC) = (BA)C = IC = C

29
Return to index
 INVERTIBLE MATRICES

a b 
A 2 x 2 matrix A  is invertible if and only if:
 c d 

ad  bc  0

In this case, the inverse is:

1 1  d  b
A 
 ad  bc   c a 

30
Return to index
 INVERTIBLE MATRICES

 d  b
Proof: Let M    and   ad  bc
 c a 
1 1  1 0
If   0 then: MA  AM  0 1 as required.
   
If 0 then MA  AM  0
Hence, if A is invertible:
MAA 1  0A 1  M  0  a  b  c  d  0  A  0
Therefore: I  AA 1  0A 1  0
This is a contradiction and so A is not invertible.
QED

31
Return to index
 INVERTIBLE MATRICES
If A and B are invertible matrices then AB is invertible and
(AB)-1 = B-1A-1.

Proof:
(AB) (B-1A-1) = A(BB-1)A-1 = A(IA-1) = AA-1= I
Similarly: (B-1A-1) (AB) = I
QED

More generally, if A1, A2, A3,.....,Ar are all invertible

matrices then:

 A1A 2 A 3 ......A r  1  A r1A r11A r12 ......A11

32
Return to index
 INVERTIBLE MATRICES
If A is an invertible matrix then AT is invertible and:

A  T 1
 A 1 T


Proof:
Since  AA    A A   I , it follows that:
1 1

 AA    A A   I
1 T 1 T

and therefore: A  A  A A   I
1 T T T 1 T

Hence A 
1 T
is the inverse of AT .
QED

33
Return to index
 CARTESIAN DECOMPOSITION
If A is an n x n matrix then it can be written as the sum of
a symmetric matrix and an anti-symmetric matrix. The
decomposition is unique.
Proof:
A   A  A    A  AT 
1 T 1
Write
2 2

Clearly
1
 A  AT  is symmetric and
1
2
 A  AT 
2
is antisymmetric.
For uniqueness, assume that A  A sm  A asm where
A sm ,A asm are symmetric and anti-symmetric,
respectively. It is then straightforward to show that:

  A  A  , A asm   A  AT 
1 T 1
A sm
2 2 QED 34
Return to index
 ORTHOGONAL MATRICES
If A and B are n x n orthogonal matrices then AB is
orthogonal.

Proof: If A and B are orthogonal then:

AT A  AAT  I BT B  BBT  I
Hence:
 AB T AB  BT  AT A B  BT IB  BT B  I
Similarly

AB AB  T
I
QED

35
Return to index
 SOME ‘UNEXPECTED’ RESULTS

2 1  2 3 
Let A    and B    4  6
6 3  
 2 1  2 3  0 0 
Then: AB       
 6 3   4  6   0 0 
Therefore: If AB = 0 this does NOT imply that either A =
0 or B = 0.

A consequence of this is that if AB = AC, then it is not

necessary that B = C.

36
Return to index
 SOME ‘UNEXPECTED’ RESULTS

2 1  2 3 
Let A    and B    4  6
6 3  
 2 1  2 3  0 0 
Then: AB       
 6 3   4  6   0 0 

Notice also that:

 2 3  2 1  22 11 
BA       
  4  6  6 3    44 22 
Therefore: AB is NOT necessarily equal to BA.

37
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Consider the following system of m linear equations
involving n variables xi:

a11 x1  a12 x 2  a13 x3  .....  a1n x n  b1

a 21x1  a 22 x 2  a 23 x3  .....  a 2n x n  b2
a 31x1  a 32 x 2  a 33 x3  .....  a 3n x n  b3
. . . ..... . .
. . . ..... . .
a m1x1  a m 2 x 2  a m 3 x3  .....  a mn x n  bm

This may be written more precisely as:

j n

a x
j 1
ij j  bi , i  1,....., m

38
Return to index
 SYSTEMS OF LINEAR EQUATIONS
The matrix representation of this system is:

 a11 a12 a13 . . . . a1n   x1   b1 

a a22 a23 . . . . a2n  x  b 
 21  2  2
 a31 a32 a33 . . . . a3 n   x3   b3 
     
 . . . . . . . .   .   . 
 . . . . . . . .   .   . 
     
 . . . . . . . .   .   . 
a am 2 am 3 . . . . amn  x  b 
 m1  m  m

or:
Ax  b

39
Return to index
 SYSTEMS OF LINEAR EQUATIONS
The ‘augmented’ matrix for the system is defined as:

 a11 a12 a13 . . . a1n b1 

a b2 
 21 a22 a23 . . . a2n
 a31 a32 a33 . . . a3 n b3 
 
 . . . . . . . . 
 . . . . . . . . 
 
 . . . . . . . . 
a bm 
 m1 am 2 am 3 . . . amn

40
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example:

5 x  7 y  13 5 7  x  13  5 7 13
4 x  5 y  10  4 5  y   10   4 5 10
      

41
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example:

5 x  7 y  13 5 7  x  13 
4 x  5 y  10  4 5  y   10 
    
This may be written as: Ax  b where

5 7 x 13
A  , x   , b   
 4 5 y  10

Then: A 1Ax  A 1b i.e: x  A 1b

42
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example:

5 x  7 y  13 5 7  x  13 
4 x  5 y  10  4 5  y   10 
    
This may be written as: Ax  b where

5 7 x 13
A  , x   , b   
 4 5 y  10
1 1  5  7
The inverse of A is: A    
3  4 5 
Hence: 1  5  7 13  5 3
x      
3   4 5  10   2 
 3
43
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example: Augmented
matrix

5 x  7 y  13 5 7  x  13  5 7 13
4 x  5 y  10  4 5  y   10   4 5 10
      
Multiply 1st equation by 4 and 2nd equation 5:

20 x  28 y  52 20 28  x  52 20 28 52

20 x  25 y  50 20 25  y   50  20 25 50
      
Replace row2 with row2row1:

20 x  28 y  52 20 28   x   52  20 28 52 
 3y   2  0  3   y     2  0  3  2
      

44
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example: Augmented
matrix
Multiply 1st equation by 3 and 2nd by 28:

60 x  84 y  156 60 84  x  156 60 84 156 

84 y  56  0 84  y    56   0 84 56 
      
Replace row1 with row1row2:
60 x  100 60 0  x  100 60 0 100 
3y  2  0 3  y    2  0 3 2 
      
Divide row1 by 60 and row2 by 3:

x  53  1 0  x  5 3  1 0 5 
3
y  23 0 1  y    2  
     3 0 1 2 3 
 

45
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Elementary row operations:

Multiply a row by a non-zero scalar: ri  ri

Add a multiple of one row to another row: ri  ri  r j

Interchange one row with another: ri  r j

If A is an m x n matrix, then all matrices that can be

obtained from A by applying a sequence of elementary
row operations, are said to be row equivalent to A.

46
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example: Augmented
matrix

5 x  7 y  13 5 7  x  13  5 7 13
4 x  5 y  10  4 5  y   10   4 5 10
      

r1  4r1 r2  5r2
20 x  28 y  52 20 28  x  52 20 28 52
20 x  25 y  50 20 25  y   50  20 25 50
      

r2  r2  r1
20 x  28 y  52 20 28   x   52  20 28 52 
 3y   2  0  3   y     2  0  3  2
      

47
Return to index
 SYSTEMS OF LINEAR EQUATIONS
Example: Augmented
matrix
r1  3r1 r2  28r2
60 x  84 y  156 60 84  x  156 60 84 156 
84 y  56  0 
 84  y   56 
    0 84 56 
 
r1  r1  r2
60 x  100 60 0  x  100 60 0 100 
3y  2  0 
 3  y   2 
   0 3 2 
 
r1  r1 / 60 r2  r2 / 3
x  53  1 0  x  5 3  1 0 5 
3
      
y  23 0 1  y   2 3 0 1 2 3 
 

48
Return to index
 ECHELON FORM OF MATRICES

A matrix is said to be in (row) echelon form if it satisfies

the following conditions:

1. All zero rows are at the bottom of the matrix.

2. The first non-zero entry in a non-zero row is called

the leading entry for that row. For each non-zero row, the
leading entry appears below and to the right of any
leading entries in preceding rows.

49
Return to index
 ECHELON FORM OF MATRICES
A matrix is said to be in reduced (row) echelon form if it
satisfies the following conditions:

1. All zero rows are at the bottom of the matrix.

2. The first non-zero entry in a non-zero row is 1. This is

called the leading one for that row.

3. For each non-zero row, the leading one appears below

and to the right of any leading ones in preceding rows.

4.If a column contains a leading one, then all other entries in

that column are zero.

Matrices in reduced echelon for are often referred to as Hermite

matrices.
50
Return to index
 ECHELON FORM OF MATRICES
Example:

6 0 2 3 1 0 0 0 2 0 0 2 0
0 1 1 5 3 4  0 0 0 0 3 0 0 
 
A  0 0 0 3 0 0 B  0 0 0 0 0 1  3
   
0 0 0 0 1  1 0 0 0 0 0 0 4
0 0 0 0 0 0  0 0 0 0 0 0 0 

A and B are in row echelon form.

51
Return to index
 ECHELON FORM OF MATRICES
Example:

1 0 0 0 0 0 0 0 1 0 0 0 0
0 1 1 0 0 4  0 0 0 0 1 0 0
 
A  0 0 0 1 0 0 B  0 0 0 0 0 1 0
   
0 0 0 0 1  1 0 0 0 0 0 0 1
0 0 0 0 0 0  0 0 0 0 0 0 0

A and B are in reduced row echelon form. i.e A and B are

Hermite matrices

52
Return to index
 PROCEDURES FOR SOLVING SYSTEMS OF
LINEAR EQUATIONS

Gaussian elimination:

Original Augmented Equivalent Solutions by

system matrix echelon back
matrix substitution

Gauss-Jordan elimination:
Equivalent
Original Augmented Hermite Write down
system matrix matrix solutions

53
Return to index
 SOLVING SYSTEMS OF LINEAR EQUATIONS
Example: Find all solutions of the system:
x
5 x  7 y  5z  13 5 7  5   13
4 x  5 y  2z  10  4 5 2   y   10
    
z
The augmented matrix is:
5 7  5 13
4 5 2 10

r1  4r 1, r2  5 r 2 20 28  20 52
20 25 10 50

r2  r2  r1 20 28  20 52 
0  3 30  2


54
Return to index
 SOLVING SYSTEMS OF LINEAR EQUATIONS

20 28  20 52 
0  3 30  2

r1  3r1 , r2  28r2 60 84  60 156 
0  84 840  56

r1  r1  r2 60 0 780 100 
 0  84 840  56
 
 1 0 13 5 
r1  r1 / 60, r2   r2 / 84  3
0 1  10 2 3
 
The general solution is therefore:

x  13z  5 3 , y  10z  2 3
55
Return to index
 SOLVING SYSTEMS OF LINEAR EQUATIONS
Example: Find all solutions of the system:

2 x  3 y  2z  1 2 3 2  x   1
3 x  y  2z  5  3  1 2  y    5 
    
x  4y  z  4  1  4 1  z   4
2 3 2 1
The augmented 3  1 2 5 
matrix is:  
 1  4 1 4  1 0 0 16 
 11 
The equivalent Hermite matrix is: 0 1 0  7 11
 
 0 0 1 0 
Hence there is a unique solution:
x = 16/11, y = 7/11, z = 0.
56
Return to index
 SOLVING SYSTEMS OF LINEAR EQUATIONS
Example: Find all solutions of the system:

2x  3y  z  2 2 3 1  x   2
3 x  y  2z  5  3  1 2  y    5 
    
x  4y  z  4  1  4 1  z   4
2 3 1 2
The augmented 3  1 2 5 
matrix is:  
 1  4 1 4
1  4 1 4 
The equivalent echelon matrix is: 0 11  1  7
 
0 0 0 1 
The last row implies that 0x+0y+0z = 1. Hence there are
no solutions to this system.
57
Return to index
 SOLVING SYSTEMS OF LINEAR EQUATIONS
Example: Find all solutions of the system:

2x  3y  z  1 2 3 1  x   1
3 x  y  2z  5  3  1 2  y    5 
    
x  4y  z  4  1  4 1  z   4
2 3 1 1
The augmented 3  1 2 5 
matrix is:  
 1  4 1 4 1 0 7 16 
 11 11 
The equivalent Hermite matrix is: 0 1  111  7 11
 
 0 0 0 0 

x  7z  16 , y z  7
11 11 11 11
58
Return to index
 DETERMINANTS
Consider the general linear system involving 2
parameters, x1 and x2:

a11 x1  a12 x 2  b1  a11 a12   x1   b1 

a21x1  a22 x 2  b2 a a   x    b  Ax  b
 21 22   2   2 
The solution is easily shown to be:
a22 b1  a12 b2 a11b2  a21b1
x1  x2 
a11a22  a12a21  a11a22  a12a21 
provided a11a22  a12a21  0

The parameter a11a22  a12a21 is called the determinant

of the matrix A and is denoted by det A.

59
Return to index
 DETERMINANTS
Therefore, the 2-parameter linear system has a solution iff
det A  0. That solution is unique.

The solution is often written in the form:

b1 a12 a11 b1
b2 a22 a21 b2
x1  x2 
a11 a12 a11 a12
a21 a22 a21 a22

1 2
or: x1  , x 2 
0 0
This is known as Cramer’s rule.

60
Return to index
 DETERMINANTS
Example: 3x  y  4
4 x  3y  2

3 1 4 1 3 4
0  5 1   10 2   10
4 3 2 3 4 2

Hence:
1 2
x  2, y   2
0 0

61
Return to index
 DETERMINANTS

Consider the general linear system involving 3

parameters, x1, x2 and x3:

a11 x1  a12 x2  a13 x3  b1

a21x1  a22 x2  a23 x3  b2
a31x1  a32 x 2  a33 x3  b3

a11 a12 a13   x1   b1 

i.e: a a a   x   b  or: Ax  b
 21 22 23   2   2 
a31 a32 a33   x3  b3 

62
Return to index
 DETERMINANTS

Eliminating x2 and x3:

x1 
 a22a33  a23a32  b1   a12a33  a13a32  b2   a12a23  a22a13  b3
a11  a22a33  a23a32   a12  a21a33  a23a31   a13  a21a32  a22a31 

We define the denominator to be the determinant of the

matrix A:

det A  a11  a22a33  a23a32 

 a12  a21a33  a23a31 
 a13  a21a32  a22a31 

63
Return to index
 DETERMINANTS
Then, if det A  0, the solution to the 3-parameter system
is given by:
1 2 3
x1  , x 2  , x3 
0 0 0
where:
a11 a12 a13 b1 a12 a13
0  a21 a22 a23 1  b2 a22 a23
a31 a32 a33 b3 a32 a33

a11 b1 a13 a11 a12 b1

2  a21 b2 a23 3  a21 a22 b2
a31 b3 a33 a31 a32 b3

64
Return to index
 DETERMINANTS
Note:

0 is the determinant of the coefficient matrix A.

To determine 1, start with the matrix A; replace column 1

with the column b and then evaluate the determinant.
To determine 2, start with the matrix A; replace column 2
with the column b and then evaluate the determinant.
Similarly for 3.

65
Return to index
 DETERMINANTS
Example: 2 x1  x 2  x3  1
x1  2 x 2  x3  2
4 x1  5 x 2  2 x3  0

2 1 1 1 1 1
0  1 2  1  5 1  2 2  1  5
4 5 2 0 5 2

2 1 1 2 1 1
2  1 2  1  10 3  1 2 2  15
4 0 2 4 5 0

The solution is therefore:

x1  1, x 2  2, x3  3
66
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

Let A be an n x n matrix. The i,j -th minor of A is the (n-1)

x (n-1) matrix formed by removing the i-th row and j-th
column of A.

Example: 1 2 0 1
3 2  2 0 
A
1 0 0 1
 
 1 2 1 0
The 2,3 minor of A is:

 1 2  1
M 23  A    1 0 1 
 1 2 0 

67
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

Let A be an n x n matrix. The i,j -th cofactor of A is given

by:
c ij  A     1 det Mij  A 
i j

Example: 1 2  10
3 2  2 0 
A
1 0 0 1
 
 1 2 1 0
The 2,3 cofactor of A is:

1 2 1
  
c 23 A   1 5
1 0 1  6
1 2 0
68
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

Let A be an n x n matrix. The determinant of A may be

written as by:

n
det A   aik c ik  ai 1c i 1  ai 2c i 2  .....  ain c in
k 1

This is an expansion by the i-th row.

(Note that this formula is a recurrence relation.)

The above formula may be written as:  det A  I  AcT

1 1
from which A  cT
det A
69
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

Example  2 3 2
A  3  1 2
 1  4 1
The minors are:
1 2 3 2 3  1
M11   M12   M13   
 4 1 1 1  1  4 
3 2 2 2 2 3 
M21   M22   M23   
 4 1 1 1  1  4 
 3 2 2 2 2 3 
M31    M32   M33   
  1 2  3 2  3  1

70
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

Example  2 3 2
A  3  1 2
 1  4 1
The corresponding cofactors are:
c11  7, c12  1, c13  11
c 21  11, c 22  0, c 23  11
c31  8, c32  2, c33  11

Hence, det A is (using rows 1, 2 and 3, respectively):

det A  a11c11  a12c12  a13c13   2  7    3  1   2  11  11

det A  a21c 21  a22c 22  a23c 23   3  11    1 0    2  11  11
det A  a31c31  a32c32  a33c33   1 8     4  2   1 11  11

71
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

The Adjoint of an n x n matrix is the transpose of the

matrix of cofactors.

For the previous example:

 2 3 2
A  3  1 2
 1  4 1
The matrix of cofactors is:  7  1  11
c ij    11 0 11 
 8 2  11

The adjoint of A is:  7  11 8 

adj A    1 0 2 
  11 11  11
72
Return to index
 LAPLACE EXPANSION OF DETERMINANTS

 7  11 8 
Hence: 1  1 
A   1 0 2
11  
  11 11  11

To confirm this:

2 3 2  7  11 8   1 0 0
1  1   1   0 1 0 
AA  3  1 2 0 2
11     
 1  4 1   11 11  11 0 0 1

73
Return to index
 GENERAL PROPERTIES OF DETERMINANTS

Let A and B be n x n matrices and let I denote the n x n

identity matrix .
Then:
det I = 1.
det A = 0, if A has a row of zeros.
det B = det A, if B is obtained from A by an elementary
row operation of the form ri  ri .
det B = det B, if B is obtained from A by an elementary
row operation of the form ri  rj.

74
Return to index
 GENERAL PROPERTIES OF DETERMINANTS

Let A and B be n x n matrices and let I denote the n x n

identity matrix .
Then:
det A = 0 if and only if A does not have an inverse .
det A  0, if and only if is invertible.
det AB = det A det B
det (A) = n det A
det (I) = n
det AT = det A
If A is an upper (or lower) triangular matrix then det A is
the product of the entries on the main diagonal.

75
Return to index
 DETERMINING THE INVERSE USING GAUSS-
JORDAN ELIMINATION

Let A be n x n matrix. If A is invertible (i.e non-singular)

then it is row equivalent to the identity matrix In. Indeed,
the inverse is also row equivalent to the identity matrix In.
Form a partitioned matrix by placing In alongside A:
A In 

Now, if Gauss-Jordan elimination is applied until the left-

hand partition becomes In , then the right-hand partition
will become the inverse of A.

A 
In   In A 1 

76
Return to index
 DETERMINING THE INVERSE USING GAUSS-
JORDAN ELIMINATION
1 0 2 
Example: Find the inverse of: A  0 1  1
 1 2  1
det A = 1 ( 0) and so the inverse exists.

Form the following 3 x 6 matrix:.

 1 0 2 1 0 0
 A In   0 1  1 0 1 0
 1 2  1 0 0 1

77
Return to index
 DETERMINING THE INVERSE USING GAUSS-
JORDAN ELIMINATION
Apply the following reduction:
1 0 2 1 0 0
r3  r3  r1 0 1  1 0 1 0 
 
0 2  3  1 0 1

 1 0 2 1 0 0
r3  r3  2r2 0 1  1 0 1 0 
 
0 0  1  1  2 1

1 0 0  1  4 2 
r1  r1  2r3 0 1 0 1 3  1
r2  r2  r3  
0 0  1  1  2 1 
78
Return to index
 DETERMINING THE INVERSE USING GAUSS-
JORDAN ELIMINATION
Finally:
1 0 0  1  4 2 
r3   r3 0 1 0 1 3  1
 
0 0 1 1 2  1

This has the form: I n A 1 

 1  4 2 
A 1   1 3  1
and so

 1 2  1
 1 0 0
It is easily confirmed that: AA 1  A 1A  0 1 0
0 0 1
79
Return to index