MAT 224

Dr. Alao.
Solution of nonlinear equations
Transcendental Equation: a polynomial containing log, trig or exp

A transcendental equation is an equation that involves transcendental functions, which are functions
that cannot be expressed as a finite combination of algebraic functions (i.e., polynomials) and
elementary functions.

Steps in Problem Solving

 Problem description
 Problem Formulation
 Problem Solution
 Using the solution

General Iteration Method

 ℜ−arrange f ( x )=0 , ¿ form x=g ( x ) { iteration function }
 ∋ x n +1=g ( x n ) ,n ≥ 0
 There are at least n+1 iteration functions for a polynomial of degree n.
 General Stopping Criterion
o For accuracy ¿ k decimal places , continue theiteration until :
1
|x n+ 1−x n|< 2 10−k (k=3 for this course , usually)
 Do not divide through by x
 Condition for Convergence
o | '
|
If g ( x 0 ) <1 , thenthe iteration method converges

Regula-Falsi Method
x ¿ f ( x n ) −x n f ( x ¿ )
x n+1= f ( x )=2 x 3−7 x +2=0 f ( 0 )=2 , f ( 1 )=−3 → x¿ =0 , x n=1
f ( x n) −f ( x ¿ )
0−2 0−0.4 ( 2 ) 0.8
x 1= =0.4 , f ( x 1 )=−0.672 x 2= = =0.2994 , f ( x2 ) =−0.0421
−3−2 −0.672−2 2.672
−0.2994 ( 2 )
x 3= =0.29322 , f ( x 3 ) =−0.002168
−0.0421−2
−0.29322 ( 2 ) −0.2929 ( 2 )
x4= =0.2929 , f ( x 4 )=−0.00006015x 5= =0.29289x ≈ 0.29289
−0.002168−2 −0.0006015−2

−1 −0.5 ( 1 ) −0.3636
x ¿=0 , x 1=1 x 1= =0.5 x 2= =0.3636 x 3= =0.3487¿ so on
−1−1 −0.375−1 −0.0427−1
NEWTON-RAPHSON
 f ( xn)
x n+1=x n − '
, n≥ 0
f ( xn )

Example:

f ( x )=x 3−5 x +1f ' ( x )=3 x 2−5 f ( 1 ) =8−10+ 1=−1 , f ( 0 )=1let x 0=0.5
( 0.5 )3−5 ( 0.5 ) +1 1.375
→ x1=0.5− =0.5+ =0.17647
2
3 ( 0.5 ) −5 −4.25
3
0.17647 −5 ( 0.17647 )+ 1 0.12314
x 2=0.17647− =0.17647− ¿ 0.20157
3 ( 0.17647 ) −5 2
−4.9066
( 0.20157 )3−5 ( 0.20157 ) +1
x 3=0.20157− 2
=0.20164
3 ( 0.20157 ) −5
( 0.20164 )3−5 ( 0.20164 ) +1
x 4 =0.20164− 2
≈ 0.20164 → x ≈ 0.20164
3 ( 0.20164 ) −5

0.5
0.5 e −cos 0.5
f ( x )=x e −cos x → f ( x )=e + x e +sin x let x 0=0.5→ x1=0.5−
x ' x x
0.5 0.5
=0.5707
e +0.5 e +sin 0.5
x 2=0.567149x 3=0.567126

SOLUTION OF LINEAR SYSTEMS OF EQUATIONS

Direct Method

Iterative Method
GAUSS-JACOBI

given a system of equations :a 11 x 1 +a 12 x 2+ a13 x 3=b 1a 21 x 1+ a22 x 2 +a23 x3 =b2

a 31 x 1+ a32 x 2+ a33 x 3=b3 Assuming that aii ≠ 0
1
a 11 x 1=b1−a 12 x 2−a13 x3 → ( x1 ) k+1= ( b −a ( x ) −a ( x ) )
a11 1 12 2 k 13 3 k
1
a 22 x 2=b2−a21 x 1−a23 x 3 → ( x 2 )k+ 1= ( b 2−a21 ( x1 ) k −a23 ( x 3 )k )
a22
1
a 33 x 3=b3 −a32 x 2−a31 x 1 → ( x 3 )k +1= ( b 3−a 32 ( x 2 )k −a 31 ( x 1 )k )
a33

Example :4 x1 + x 2 + x 3=2x 1+ 5 x 2 +2 x3 =−6x 1+ 2 x 2 +3 x3 =−4with ( x n )0=0

1 1 1
→ ( x 1) k+1=
4
( 2−( x 2 )k −( x 3 ) k )→ ( x 2) k+1= ( −6−( x 1) k −2 ( x 3 )k )→ ( x 3 )k+1= ( −4−2 ( x 2 )k −( x1 ) k )
5 3
 1 1
k =0→ ( x 1) 1=
4
( 2−( x 2 )0 −( x 3 )0 )=0.5→ ( x 2) 1= (−6−( x 1 )0−2 ( x 3 )0 ) =−1.2
5
1 −4
→ ( x 3 )1= (−4−2 ( x 2 )0 −( x 1 )0 )=
3 3

1
k =1→ ( x 1) 2=
4
(
1 6 4 17
2−( x2 )1− ( x3 ) 1) = 2+ + =
4 5 3 15( )
1
5
1
5
8
(
→ ( x 2) 2= (−6−( x 1 )1−2 ( x 3 )1 )= −6+ −0.5 =¿
3 )
1 1
→ ( x 3 )2= (−4−2 ( x2 )1− ( x1 )1 ) = −4 +
3 3 (2 (6 )
5
−0.5 =¿ )
1
k =2→ ( x 1) 2=
4
(
1 6 4 17
2−( x2 )1− ( x3 ) 1) = 2+ + =
4 5 3 15 ( )
1
5
1
5
8
(
→ ( x 2) 2= (−6−( x 1 )1−2 ( x 3 )1 )= −6+ −0.5 =¿
3 )
→ ( x 3 )2=
1
3 (
1
(
−4−2 ( x2 )1− ( x1 )1 ) = −4 +
3
2 (6 )
5
−0.5 =¿ )
1
k =3→ ( x 1) 2=
4
(
1 6 4 17
2−( x2 )1− ( x3 ) 1) = 2+ + =
4 5 3 15( )
1
5
1
5
8
(
→ ( x 2) 2= (−6−( x 1 )1−2 ( x 3 )1 )= −6+ −0.5 =¿
3 )
1 1
→ ( x 3 )2= (−4−2 ( x2 )1− ( x1 )1 ) = −4 +
3 3 (2 (6 )
5
−0.5 =¿ )
1
k =4 → ( x 1) 2=
4
(
1 6 4 17
2−( x2 )1− ( x3 ) 1) = 2+ + =
4 5 3 15 ( )
1
5
1
5
8
(
→ ( x 2) 2= (−6−( x 1 )1−2 ( x 3 )1 )= −6+ −0.5 =¿
3 )
→ ( x 3 )2=
1
3 (
1
(
−4−2 ( x2 )1− ( x1 )1 ) = −4 +
3
2 (6 )
5
−0.5 =¿ )
1
k =5→ ( x 1) 2=
4
(
1 6 4 17
2−( x2 )1− ( x3 ) 1) = 2+ + =
4 5 3 15( )
1
5
1
5
8
(
→ ( x 2) 2= (−6−( x 1 )1−2 ( x 3 )1 )= −6+ −0.5 =¿
3 )
1 1
→ ( x 3 )2= (−4−2 ( x2 )1− ( x1 )1 ) = −4 +
3 3 (2 (6 )
5
−0.5 =¿ )
Example 2
 Solvethe systemof eqn :26 x +2 y+ 2 z=12.63 x+ 27 y + z=−14.32 x+3 y +17 z=6
1 1 1
→ x k+1=
26
( 12.6−2 y k −2 z k )→ y k+1= (−14.3−z k −3 x k ) → z k+1= (6−2 x k −3 y k )
27 17

1 12.6 1 −14.3
k =0 x 1=
26
( 12.6−2 y 0−2 z 0 )=
26
y 1= ( −14.3−z 0−3 x 0 ) =
27 27
1 6
z 1= ( 6−2 x 0−3 y 0 )=
17 17

(
k =1x 2=
1
26
12.6−2
27(
−14.3
) ( ))
−2
6
17
=0.4982

y 2=
1
27(−14.3−
6
17 ( ) ( ))
−3
12.6
26
=−0.596547z =
17 (
1
6−2 (
2
26 ) ( 27 ))
12.6
−3
−14.3
=0.38939

1 1
k =2 x 3= ( 12.6−2 y 1−2 z 1) = ( 12.6−2 (−0.596547 ) −2 ( 0.38939 ) )=0.50055
26 26
1 1
y 3= (−14.3−z 1−3 x 1 )= (−14.3−( 0.38939 )−3 ( 0.4982 ) )=−0.5994
27 27
1 1
z 3= ( 6−2 x 1−3 y 1 )= ( 6−2 ( 0.4982 )−3 (−0.596547 ) )=0.3996
17 17

1 1
k =3 x 4 = ( 12.6−2 y 3 −2 z 3 ) = ( 12.6−2 (−0.5994 )−2 ( 0.38939 ) )=0.49998
26 26
1 1
y 4 = (−14.3−z 3−3 x 3 )= (−14.3−( 0.39960 )−3 ( 0.500550 ) )=−0.6
27 27
1 1
z 4 = ( 6−2 x 3−3 y 3 ) = ( 6−2 ( 0.500550 )−3 (−0.5994 ) )=0.3998
17 17

1 1
k =4 x 5= ( 12.6−2 y 4 −2 z 4 )= ( 12.6−2 (−0.6 )−2 ( 0.3998 ) )=0.50001
26 26
1 1
y 5= (−14.3−z 4 −3 x 4 )= (−14.3− ( 0.3998 )−3 ( 0.49998 ) )=−0.59999
27 27
1 1
z 5= ( 6−2 x 4−3 y 4 )= ( 6−2 ( 0.49998 )−3 (−0.6 ) ) =0.4
17 17

x 1 ≈ 0.5 , x 2 ≈−0.6 , x 3 ≈ 0.4

GAUSS-SEIDAL
 given a system of equations :a 11 x 1 +a 12 x 2+ a13 x 3=b 1a 21 x 1+ a22 x 2 +a23 x3 =b2
a 31 x 1+ a32 x 2+ a33 x 3=b3 Assuming that aii ≠ 0
1
a 11 x 1=b1−a 12 x 2−a13 x3 → ( x1 ) k+1= ( b −a ( x ) −a ( x ) )
a11 1 12 2 k 13 3 k
1
a 22 x 2=b2−a21 x 1−a23 x 3 → ( x 2 )k+ 1= ( b 2−a21 ( x1 ) k+1−a23 ( x 3 )k )
a22
1
a 33 x 3=b3 −a32 x 2−a31 x 1 → ( x 3 )k +1= ( b 3−a 32 ( x 2 )k+1−a31 ( x 1 )k +1)
a33

Example

45 x +2 y +3 z=58−3 x+ 22 y +2 z=475 x+ y+ 20 z =67.using x i=0

1 1 1
→ x k+1= ( 58−2 y k −3 z k ) , y k+ 1= ( 47−2 z k+1 +3 x k ) , z k+1 = ( 67−5 x k +1− y k +1 )k =0
45 22 20
58
x 1= , y 1=
45
1
22 (
47−0+ 3
58
45( ))
=2.31212, z 1=
1
20 ( 58
67− −2.31212 =2.91217
9 )
1
( 58−2 ( 2.31212 )−3 ( 2.91217 ) )=0.99198 ,
k =1 x 2=
45
1 1
y 2= ( 47−2(2.91217)+ 3 ( 0.99198 ) )=2.00689 , z 2= ( 67−5(0.99198)−2.00689 )=3.00166
22 20

1
( 58−2 ( 2.00689 )−3 ( 3.00166 ) )=0.99958 ,
k =2 x 3=
45
1 1
y 3= ( 47−2(2.91217)+3 ( 0.99198 ) )=1.99979 , z 3= ( 67−5(0.99198)−2.00689 )=3.0020155
22 20

ASSIGNMENT

Gx +3 y+ 52 z=173.61x−27 y +2 z=71.3141 x−2 y +3 z=65.46 x 0=1 , y 0=−1 , z 0=3

Convergence test :diagonal dominance|aii|>|aij|+|aik|+ … ( ∀ i ≠ j ,i ≠ k , … )

for i=3 ,|a33|>|a31|+|a 32|→|3|>|41|+|−2| ,WHICH IS FALSE
→ we cannot guarantee convergence of theGauss−Seidel∨Jacobi method

−1
SEIDAL x k+1 =( 173.61−3 y k −52 z k ) y k +1= ( 71.31−x k +1−2 z k )
27
1
z k+1 = ( 65.46−41 x k+1 +2 y k+1 )
3
 −1
k =0x 1=( 173.61+3−52 ( 3 ) )=20.61 y 1= ( 71.31−20.61−6 )=−1.65556
27
1
z 1= ( 65.56−41 ( 20.61 )−2 ( 1.65556 ) )=−260.9537 k =1
3
x 2=( 173.61−52 (−260.9537 )+ 3 (1.65556 ) )=13748.1693
−1
y 2= ( 71.31−2.0 (−260.9537 )−1.0 ( 13748.1693 )) =487.2204
27
1
z 2= ( 65.46−41.0 ( 13748.1693 )+ 2.0 ( 487.2204 )) =−187545.0129
3
the values of x , y∧z tend ¿ ∞

−1 1
JACOBI x k+1 =( 173.61−3 y k −52 z k ) y k +1= ( 71.31−x k −2 z k ) z k+1 = ( 65.46−41 x k + 2 y k )
27 3
−1
k =0x 1=( 173.61+3−52 ( 3 ) )=20.61 y 1= ( 71.31−20.61−6 )=−1.65556
27
1
z 1= ( 65.56−41 ( 20.61 )−2 ( 1.65556 ) )=−260.9537 k =1
3
x 2=( 173.61−52 (−260.9537 )+ 3 (1.65556 ) )=13748.1693
−1
y 2= ( 71.31−2.0 (−260.9537 )−1.0 ( 13748.1693 )) =487.2204
27
1
z 2= ( 65.46−41.0 ( 13748.1693 )+ 2.0 ( 487.2204 )) =−187545.0129
3
the values of x , y∧z tend ¿ ∞

Decision variables

Constraints

Objective Function

Maximise Z = 23x+16y

Maximise Z=12 x+16 y ( objective function )10 x+ 20 y ≤ 120 , 8 x +8 y ≤ 80 , x , y ≥ 0Constraint 1

at x=0 →20 y =120→ y=6at y=0 → 10 x=120 → x=12→ ( 0 , 6 ) , ( 12 , 0 ) Constraint 2
at x=0 →8 y =80 → y =10at y=0 → 8 x =80 → x=10 ( 10 , 0 ) ,(0 , 10)
A ( 0 , 0 ) , B ( 10 , 0 ) , C ( 8 , 2 ) , D ( 0 , 6 ) Z ( A )=0 , Z ( B )=120 , Z ( C ) =96+32=128 , Z ( D )=96
→ Z ismaximised at C=(8 ,2)

Z=50 x+18 y2 x+ y ≤ 100x + y ≤ 80x , y ≥ 0First constraint x=0 : y =100 , ( 0,100 )

y=0: x =50 , ( 50 , 0 ) 2 nd constraint x=0 : y =80 , ( 0 ,80 ) y=0: x =80 , ( 80 ,0 )
A ( 0 , 0 ) , B ( 0 , 80 ) , C ( 20 , 60 ) , D ( 50 , 0 ) Z ( A )=0 , Z ( B )=1440 , Z ( C ) =2080 , Z ( D )=2500
→ Z ismaximised at D
Dr. O.
INTERPOLATION & APPROXIMATION
 Show workings in detail
 Exact value is obtained when using differential equation methods
 Approximate values are from numerical methods

Interpolating Polynomial : Pn ( x )=l 0 ( x ) f ( x 0 ) +l 1 ( x ) f ( x 1 ) +…+ l n ( x ) f ( x n )

f i∧l i ( x ) , i=0 , 1, 2 , … n , are poly of degree n Step function :l i ( x j )= [

0 , i≠ j INTERPOLATI
1 , i= j
ON Uneven intervals:
Lagrange Interpolation
1. Linear interpolation
For n=1 , we have the data : x → f ( x ) , x 0 → f 0 , x 1 → f 1
x−x 1 x −x 0
Lagrange fundamental polynomial :l 0= ,l 1=
x 0−x 1 x 1−x 0
The Lag lininterpolation is: P 1 ( x )=l 0 f 0 +l 1 f 1
2. Quadratic interpolation For n=2 : x → f ( x ) , x 0 → f 0 , x 1 → f 1 , x 2 → f 2
( x −x1 ) ( x−x 2 ) ( x−x 0 ) ( x−x 2 ) ( x−x 0 ) ( x−x 1 )
LFP :l 0 = ,l 1= , l 2=
( x 0 −x1 ) ( x 0−x 2 ) ( x 1−x 0 ) ( x 1−x 2) ( x 2−x 0 ) ( x 2−x 1 )
P2 ( x ) =l 0 f 0 +l 1 f 1+ l 2 f 2
3. Cubic interpolation
For n=2 : x → f ( x ) , x 0 → f 0 , x 1 → f 1 , x 2 → f 2 , x 3 → f 3
( x−x 1 ) ( x−x 2 )( x −x3 ) ( x−x 0 ) ( x−x 2 ) ( x−x 3 ) ( x−x 0 ) ( x−x 1 ) ( x−x 3 ) ( x−
LFP :l 0 = , l 1= , l 2= ,l 3=
( x 0 −x1 ) ( x 0−x 2 ) ( x 0−x 3 ) ( x 1−x 0 )( x1 −x2 ) ( x 1−x 3 ) ( x 2−x 0 ) ( x2 −x1 ) ( x 2−x 3 ) ( x 3−x
P3 ( x )=l 0 f 0 +l 1 f 1 +l 2 f 2+ l3 f 3
Error of interpolation
E ( f , x )=|f ( x )−Pn ( x )| {exact −¿ }

E . g .:Using sin ( 0.1 )=0.099683 ,∧sin ( 0.2 )=0.19867 , find

an≈value of sin ( 0.15 ) by LI .Obtain a bound on the error at x =0.15
Sol :data values of n → poly degree of n−1
x−x 1 x−x 0 x−0.2 x−0.1
P1 ( x ) =l 0 f 0 +l 1 f 1= f 0+ f = ( 0.099683 )+ ( 0.19867 )
x 0−x 1 x 1−x 0 1 0.1−0.2 0.2−0.1
1
¿ ( 0.099683 ( 0.2−x )+ ( x−0.1 )( 0.19867 ) )=10 (0.098807 x+ 0.0001056)
0.1
P1 ( 0.15 )=10 ( 0.098807 ( 0.15 )+ 0.0001056 ) =0.1492665
 E . g .:Using Lagrange find the quad poly that takesthe valuesx → y :0 →0 , 1 →1 , 3 →0
( x−x 0 ) ( x−x 2 ) x ( x−3 ) 1
Sol : P2=l 0 f 0+ l 1 f 1+l 2 f 2=l 1 ( 0 ) + ( 1 ) +l 2 ( 0 )¿ = ( 3 x−x 2 )
( x 1−x 0 ) ( x 1−x 2) 1 ( 1−3 ) 2
1 2
→ LQP :f ( x )= (3 x −x )
2

E . g .:Given that f ( 0 )=1 , f ( 1 )=3 , f ( 3 ) =55 ,

find the uniqe poly of degree 2∨less , which fitsthe givendata
( x−1 )( x−3 ) ( x ) ( x−3 ) x ( x−1 )
Sol : P2=l 0 f 0+ l 1 f 1+l 2 f 2= ( 1 )+ ( 3 )+ ( 55 )
(−1 )(−3 ) ( 1 )(−2 ) ( 3 )( 2 )
1 2 3 55 2
¿ ( x −4 x+ 3 ) + ( 3 x−x ) + ( x −x )=1+ x
3 2
2
6
−4 9 55
+ −
3 2 6
2 1 3 7 2
+ x − ¿ 1−6 x− x
3 2 6 ( ) ( )

E . g : f ( x )=sin x+ cos x . Find f ( 12π ) with LIP , comparing withexact

π 0
f ( 10 0 )=1.1585 , f ( 20 0 )=1.2817 , f ( 300 )=1.3660 Sol : =15
12
( x−20 0 ) ( x−300 ) ( x−100 ) ( x−300 ) ( x−200 ) ( x−100 )
P2=l 0 f 0+l 1 f 1 +l 2 f 2= ( 1.1585 ) + ( 1.2817 ) + ( 1.3660 )
(−10 0 )( −200 ) ( −100 ) ( 100 ) ( 200 ) ( 100 )
1
¿ 0
¿
200

Divided Differences
f i+1−f i f 1 −f 0
1 st order : f [ x i , x i+1 ]= , i=0 , 1 ,2 , … , n−1→ f [ x 0 , x 1 ]=
x i+1−x i x1 −x 0
f [ x i+ 1 , x i+2 ] −f [ x i , xi +1 ]
2 nd order : f [ x i , xi +1 , x i+2 ]= , i=0 , 1, 2 , … , n−1
x i+2−x i
f [ x 1 , x 2 ]−f [ x 0 , x1 ]
→ f [ x 0 , x 1 , x 2 ]=
x 2−x 0
=
1
(
f 2−f 1 f 1−f 0
−
x 2−x 0 x 2−x 1 x 1−x 0 )
1
E . g .: Find 2 nd÷diff of f ( x )= , using a , b , c
x
f [ a ,b ] =
b−a [ ]
f ( b ) −f ( a )
=
1 1 1
− =
b−a b a ( b−a ) ab ab
a−b
=
−1

f [b , c ]=
f c −f b
[ ]
=
1 1 1
c−b c−b c b bc ( c−b ) bc
− =
b−c
=
−1

f [ a ,b , c ] =
f [ b , c ] −f [a , b]
c−a
=
[ ]
1 −1 1
( ) + =
1
c−a bc ab b ( c−a ) a c ( c−a ) abc abc
1 1
− =
c −a
=
1
 f 0−f 1 −8−3
E . g .:Obtain÷diff table f (−1 )=−8 , f ( 0 )=3 , f ( 2 )=1 , f ( 3 )=12f [ x 0 , x1 ] = = =11
x 0−x 1 −1−0
f 1−f 2 3−1 f 2−f 3 1−12
f [ x 1 , x 2 ]= = =−1f [ x 2 , x 3 ] = = =11
x 1−x 2 0−2 x 2−x 2 2−3

Newton’s Divided Difference Interpolation

f ( x )=P n ( x )=c0 + ( x −x 0 ) c1 + ( x−x 0 ) ( x−x 1 ) c 2+ …+ ( x−x 0 ) ( x−x 1 ) c n f 0=c 0
f 1−c0 f 1−f 0
f 1=c 0 + ( x 1−x 0 ) c 1 → c 1= = =f [ x 0 , x 1 ]
x 1−x 0 x 1−x 0
f 2 −f 0 −( x 2−x 0 ) f [ x 0 , x 1 ]
f 2=c 0 + ( x 2−x 0 ) c 1+ ( x2 −x 0 )( x 2−x 1 ) c 2 → c 2= =f [ x 0 , x 1 , x 2 ]
( x 2−x 0 ) ( x 2−x 1 )
→ c n=f [ x 0 , x1 , x2 , … , x n ]

→ P n ( x )=f 0 + ( x−x 0 ) f [ x 0 , x1 ] + ( x−x 0 ) ( x−x 1 ) f [ x 0 , x 1 , x 2 ] +…+ ( x−x 0 )( x −x1 ) … ( x−x n−1 ) f [ x 0 , x 1 , … , x n ]

EXAMPLE

f (−4 )=1245 , f (−1 )=33 , f ( 0 )=5 , f ( 2 )=9 , f ( 5 )=1335NDDF for 4 :

P4 ( x )=f 0 + ( x−x 0 ) f [ x 0 , x 1 ] + ( x−x 0 ) ( x−x 1 ) f [ x 0 , x 1 , x 2 ] + ( x−x 0 ) ( x−x 1 )( x−x 2) f [ x 0 , x 1 , x 2 , x 3 ]+ ( x−x 0 )( x−x 1) ( x

f ( x )=1245+ ( x+ 4 )(−404 ) + ( x +4 ) ( x+1 )( 94 ) + ( x+ 4 )( x +1 ) x (−14 ) + ( x+ 4 )( x +1 ) x ( x−2 ) ( 3 )

4 3 2
¿ 3 x −5 x +6 x −14 x +5
Example

f (−2 )=9 , f (−1 )=16 , f ( 0 )=17 , f (1 )=18 , f ( 3 )=44 , f ( 4 ) =81 Interpolate at x=0.5 , x=3.1

as 4 th∧5 th dd are 0
P ( x )=f 0+ ( x−x 0 ) f [ x 0 , x 1 ] + ( x−x 0 ) ( x−x 1) f [ x 0 , x 1 , x 2 ]+ ( x−x 0 )( x −x1 ) ( x−x 2 ) ( x−x 3 ) f [ x 0 , x 1 , x 2 , x 3 ]
¿ 9+ ( x +2 ) 7+ ( x +2 ) ( x+1 )(−3 ) + ( x +2 ) ( x+ 1 ) x ¿ x 3+ 17
3 3
Hence , f ( 0.5 )=0.5 +17=17.125f ( 3.1 ) =3.1 + 17=47.791

Example

f ( 1 ) =3 , f ( 3 )=31 , f ( 4 )=69 , f (5 )=131 , f ( 7 )=351, f ( 10 )=1011 find f ( 3.5 ) , f ( 8 ) , f ' ( 3 ) ∧f ' ' (1.5)

As 4 th dd is 0
f ( x )=f 0 + ( x−x 0 ) f [ x 0 , x1 ]+ ( x−x 0 ) ( x−x 1 ) f [ x 0 , x 1 , x 2 ] + ( x−x 0 ) ( x−x 1 ) ( x−x 2 )( x−x3 ) f [ x 0 , x 1 , x 2 , x 3 ]
¿ 3+ ( x−1 ) 14+ ( x −1 )( x−3 ) 8+ ( x−1 )( x−3 )( x−4 ) 1
2 3 2 3 ' 2 ''
¿ 3+14 x−14 +8 x −32 x+24 + x −8 x + 19 x −12¿ x + x+1→ f =3 x +1 , f =6 x
3 3 ' ''
→ f ( 3.5 )=3.5 + 3.5+1=47.375 , f ( 8 )=8 + 8+1=521f ( 3 ) =3 ( 9 ) +1=28 , f ( 1.5 )=6 ( 1.5 ) =9
Interpolation with Evenly Spaced Points
FORWARD DIFFERENCE
∆ f ( x n+ h )=f n +1∆ f n=f n +1−f n∆ 2 f n=∆ f n +1−∆ f n=f n+ 2−2 f n +1+ f n
3 2 2
∆ f n=∆ f n +1−∆ f n=f n+3−2 f n+2 + f n+1 −( f n+2−2 f n+1 +f n )¿ f n+3−3 f n+ 2+3 f n+1−f n
0
→ ∆ f n is of a similar form ¿ : ( x− y ) → ∆ f n=∑ (−1 ) a f n +k
a a

k=a
a k

k ()

BACKWARD DIFFERENCE
∇ f n=f n −f n−1∇ 2 f n=∇ f n−∇ f n−1=f n−f n−1−f n−1+ f n −2=f n −2 f n−1 +f n−2
3 2 2
∇ f n=∇ f n−∇ f n−1=f n−2 f n−1 + f n−2−( f n−1−2 f n−2+ f n−3 )¿ f n−3 f n−1+3 f n−2−f n−3
n
→ ∆a f n=∑ (−1 ) a f n−k
k=a
k

k ()
Newton’s Forward Difference Interpolation Formula
1
f [ x 0 , x1 , … , xn ]=
n
n
∆ f 0¿ Newto n' s Divided difference
n!h
2 n
∆f0 ∆ f0 ∆ f0
f ( x )=f 0 + ( x−x 0 ) + ( x −x0 ) ( x−x 1 ) 2
+…+ ( x −x 0 )( x −x1) … ( x−x n−1 ) n
h 2h n! h
( x−x 0 ) ∆ f 0 ( x−x 0 ) ( x−x 1 ) ∆2 f 2
E . g .: f 2 ( x )=f 0 + +
h 2 h2

Newton’s Backward Difference Interpolation Formula

1
f [ x n , x1 , … , xn ]=
n
n
∇ f 0¿ Newto n' s Divided difference
n!h
2 n
∇fn ∇ fn ∇ fn
f ( x )=f n + ( x−x n ) + ( x −x n) ( x− xn −1 ) 2
+…+ ( x−x n )( x−x n−1) … ( x−x 0 ) n
h 2h n!h
( x−x 0 ) ∇ f n ( x−x 0 ) ( x−x 1 ) ∇ 2 f n
E . g .: f 2 ( x )=f n + +
h 2 h2
x−x n=sh , x−x n−1=( s +1 ) h , x −xn −i =( s +i ) h
( s )( s+1 ) ∇ 2 f n ( s ) ( s +1 ) ( s+ 2 ) ∇3 f n n
∇ fn
→ f ( x )=f n +s ∇ f n+ + +…+ s ( s+1 ) … ( s+n )
2 3! n!
Example
f ( 0.1 )=1.4 , f ( 0.2 ) =1.56 , f ( 0.3 ) =1.76 , f ( 0.4 )=2 , f ( 0.5 )=2.28 h=0.1Forward difference

x f (x) ∆f ∆ f
2
∆ f
3
∆ f
4

0.1 1.4
0.2 1.56 0.16
0.3 1.76 0.2 0.04
0.4 2 0.24 0.04 0
0.5 2.28 0.28 0.04 0 0

( 0.1 ) ( 0.02 )
2
( x−x 0 ) ∆ f 0 ( x−x 0 ) ( x−x 1 ) ∆ f 0 ¿ 1.4+ ( x−0.1 ) 0.16 + ( x−0.1 )( x−0.2 ) 0.04
f ( x )=f 0 + + 2
h 2h
2
¿ 2 x + x +1.28
Backward

x f (x) ∇f ∇ f
2
∇ f
3
∇ f
4

0.1 1.4
0.2 1.56 0.16
0.3 1.76 0.2 0.04
0.4 2 0.24 0.04 0
0.5 2.28 0.28 0.04 0 0

( 0.1 ) ( 0.02 )
2
( x−x n ) ∇ f n ( x−x n ) ( x−x n−1 ) ∇ f 0 ¿ 2.28+ ( x−0.5 ) 0.28 + ( x−0.5 ) ( x−0.4 ) 0.04
f ( x )=f n + + 2
h 2h
¿ 2 x + x +1.28→ f ( 0.25 ) =1.655 , f ( 0.35 )=1.875
2

2 : x=1 , f ( 0.1 )=−1.699 , f ( 0.3 )=−1.073 , f ( 0.5 )=−0.375 , f ( 0.7 )=0.443 , f ( 0.9 )=1.429
f ( 1.1 ) =2.631 withbackward h=0.2

x f (x) ∇f 2
∇ f
3
∇ f
4
∇ f
5
∇ f
0.1 -1.699
0.3 -1.073 0.62
6
0.5 -0.375 0.69 0.072
8
0.7 0.443 0.81 0.12 0.04
8 8
0.9 1.429 0.98 0.168 0.04 0
6 8
1.1 2.631 1.20 0.216 0.04 0 0
2 8
2 3
∇fn ∇ fn ∇ fn
f ( x )=f n + ( x−x n ) + ( x −x n) ( x− xn −1 ) 2
+ ( x−x n ) ( x−x n−1 ) ( x−x n−2 ) 3
h 2h 3! h
( x−1.1 ) 1.202 ( x−1.1 ) ( x−0.9 ) 0.216 ( x−1.1 )( x−0.9 ) ( x −0.7 ) 0.048
f ( x )=2.631+ + +
0.2 2 ( 0.2 )
2
3 ( 0.2 )
3

f ( x )=2.631+ 6.01 ( x−1.1 )+2.7 ( x−1.1 )( x−0.9 )+ ( x−1.1 )( x−0.9 ) ( x −0.7 )

f ( 1 ) =2.631+6.01 (−0.1 ) +2.7 (−0.1 ) 0.1+ (−0.1 ) ( 0.1 ) (−0.3 )=2.004
NUMERICAL DIFF
'
F rom Newto n s Divided difference
2 n
∆f0 ∆ f0 ∆ f0
f ( x )=f 0 + ( x−x 0 ) + ( x −x0 ) ( x−x 1 ) 2
+…+ ( x −x 0 )( x −x1 ) … ( x−x n−1 ) n
Set x=x 0 +sh
h 2h n! h
1 2 1 n
f ( x )=f ( x 0 +sh )=f 0 + s ∆ f 0 + s ( s−1 ) ∆ f 0 + …+ s ( s−1 ) ( s−2 ) … ( s−n+1 ) ∆ f 0
2 n!
x−x 0 ds 1 df 1 df 1
s= → = = = ¿
h dx h dx h ds h

Example:
'
find y at x=1 :

x y ∆y 2
∆ y
3
∆ y
1 1
2 8 7
3 27 19 12
4 64 37 18 6
h=1 , x 0=1 → x=x 0+ sh=1+s → x=1 → s=0
dy
dx
1
h ( 1
2
1
3
1
2
1
)
→ ( 1 )= ∆ f 0− ∆ 2 f 0 + ∆ 3 f 0 =7− (12 )+ ( 6 )=3
3

NUMERICAL INTEGRATION
Trapezium/Trapezoidal Rule
' 1
¿ Newto n s Forward difference formula :f ( x )=f 0 + ( x−x 0 ) ∆ f ( x0 )
h
let x 0=a , x 1=b , h=b−a=x 1−x 0
b x1 x
∆f0 x
a 0x
( 1
h )
→ I =∫ f ( x ) dx=∫ f 0 + ( x−x 0 ) ∆ f 0 dx=f 0∫ dx +
x
h ∫ x
( x−x 0 ) dx
∆f0 ∆f 0 h h h b−a
¿ ( x 1−x 0 ) f 0+
2h
[ ( x−x 0 ) ]x =( x1 −x0 ) f 0 +
2 x
1

0
2h 1 0
2
( x −x ) =h f 0 + 2 ∆ f 0=h f o + 2 ( f 1−f 0 )= 2 [ f 1 + f 0 ]= 2 ( f ( b )+
Error
b b
b−a
R ( f , x )=∫ f ( x ) dx−I f ( x )=1 : R ( f , x )=∫ dx − ( 2 )=b−a−( b−a )=0
a a 2
b
b−a 1 1
f ( x )=x : R ( f , x )=∫ x dx− ( b+ a )= ( b2−a2 )− ( b 2−a 2) =0 f ( x )=x 2 : R ( f , x )=∫
a 2 2 2

Composite trapezium rule

b x x1 xN
b−a
h= , b=x N , a=x 0∫ f ( x ) dx=∫ f ( x ) x=∫ f ( x ) dx +…+ ∫ f ( x ) dx
N a x x x 0 N −1

h h
¿ [ ( f 0 + f 1 ) + ( f 1+ f 2 ) +…+ ( f N−1 + f N ) ]= [ f 0 +2 ( f 1+ f 2 +…+ f N−1 ) + f N ]
2 2

Example 3.11 Derive the trapezium rule using the Lagrange linear interpolating polynomial

x−b x−a x ( f ( b )−f ( a ) ) +bf ( a )−af ( b )

Lagrange : f ( x )= f ( a) + f ( b )=
a−b b−a b−a

( )
b b 2 2
1 1 b −a
I =∫ f ( x ) dx= ∫ x ( f ( b ) −f ( a ) ) +bf ( a ) −af ( b ) dx = ( f ( b ) −f ( a ) ) + ( b−a ) ( bf ( a )−af ( b ) )
a b−a a b−a 2
( b+a ) b−a
¿ ( f ( b )−f ( a ) ) +bf ( a )−af ( b )= ( f ( b) + f ( a) )
2 2

1
dx
Example 3.12 Find the approximate value of I =∫ , using the trapezium rule with 2, 4 and 8
0 1+ x
equal subintervals. Using the exact solution, find the absolute errors.

b−a 1 1
Taking a=0 , b=1h= = , nodes are N +1N=2: , nodes : 0 ,0.5 ,1
N N 2
1 1
N=4 : , 0 , 0.25 , 0.5 , 0.75 ,1N=8: , 0 , 0.125 ,0.25 , 0.375 , 0.5 , 0.675 , 0.75 , 0.875 ,1
4 8
1
f ( x )=
1+ x
N=2, x 0 0.5 1
f (x) 1 0.667 0.5
 N=4 , x 0 0.25 0.5 0.75 1
f (x) 1 0.8 0.667 0.5714 0.5

N=8 , x 0 0.125 0.25 0.5 0.75 0.875 1

f (x) 1 0.8889 0.8 0.667 0.5714 0.5333 0.5

h
N=2: I 1= ( f 0 +2 f 1 + f 2 )=0.25 [ 1+2 ( 0.667 )+ 0.5 ] =0.708334
2
h
N=4 : I 2= ( f 0 +2 f 1 +2 f 2+2 f 3+ f 4 )=0.125 [ 1.0+2 {0.8+ 0.667+0.571429 }+0.5 ] =0.697024
2
h
N=8: I 3= ( f 0+ 2 f 1+ 2 f 2 +2 f 3 +2 f 4 +2 f 5 +2 f 6 + 2 f 7 +f 8 )=0.0625 [ 1.0+2 { 0.888889+0.8+ 0.727273+0.666667
2
The exact value of the integral is I =ln2=0.693147The errors ∈the solutions are the following :
|Exact – I 1|=|0.693147 – 0.708334|=0.015187|Exact – I 2|=|0.693147 – 0.697024|=0.003877
|Exact – I 3|=|0.693147 – 0.694122|=0.000975

2
dx
Evaluate I = ∫ with 4 and 8 subintervals using the trapezium rule. Compare with the exact
1 5+3 x
solution and find the absolute errors in the solutions. Comment on the magnitudes of the errors
obtained. Find the bound on the errors

b−a 1 1
h= = N=4 :h= . The nodes are 1 , 1.25 ,1.5 , 1.75 , 2
N N 4
1
N=8: h= .nodes are 1 ,1.125 , 1.25 , 1.375 ,1.5 , 1.675 , 1.75 ,1.875 , 2.0
8
N=4 , x 1 1.25 1.5 1.75 2
f (x) 0.12 0.114286 0.105263 0.101266 0.094118
5

N=8 , x 1 1.25 1.375 1.5 1.75 1.875 2

f (x) 0.125 0.114286 0.109589 0.105263 0.101266 0.097561 0.094118

Simpson’s 1/3 rule

1 1 2
f ( x )=f 0 + ( x −x 0 ) ∆ f 0 + 2 ( x−x 0 ) ( x−x 1 ) ∆ f 0
h 2h

h
I = ( f 0+ 4 f 1+ f 2)→
3
b−a
6 [
f ( a )+ 4 f
a+b
2 ( )
+f 2
]
Composite Simpson’s 1/3 rule
h
I = ( f 0 + 4 { f 1+ f 3 +…+ f 2 N−1 } +…+2 { f 2 +f 4 +…+ f 2 N −2 }+ f 2 N )
3

Least Squares Methods

Straight-Line Method
Derivation:
y ≈ y λ =a+ b x1 , … , y λ =a +b x ne 1=| y 1− y λ | ,… , e n= y n − y λ = y n −a−b x n
1 n 1 n

n n
∂S ∂S
∑ of squares , S=∑ e 2i =∑ ( y i−a−b x i )2 By principles of least squares : ∂ a = ∂ b =0
i=1 i=1

∂S ∂S 2
→ =−2 ( y i−a−b x i ) =0 → y i =a+b x i→ → =−2 xi ( y i−a−b x i ) =0 → x i yi =a x i+ b xi
∂a ∂b
2
→ Σ y=na+b Σ x , Σ xy =a Σ x +b Σ x
Parabola Method
Derivation: [Must be done in exam]
n n

Let y =a+bx +c x y λ =a+b x n +c xn e n= y n −( a+ b x n+ c x n )→ S=∑ ei =∑ ( y i−a−b x i−c x i )

2 2 2 2 2 2
n
i=1 i=1

∂ S ∂S ∂S ∂S
=−2 ( yi −a−b x i−c x i ) =0 → Σ y=na+ b Σ x +c Σ x
2 2
Principle: = = =0→
∂a ∂b ∂c ∂a
∂S
=−2 x i ( y i−a−b x i−c xi ) =0 → Σ xy=a Σ x +b Σ x+ c Σ x
2 2
→
∂b
∂S
=−2 x i ( y i−a−b x i−c x i )=0→ Σ x y=a Σ x + b Σ x +c Σ x
2 2 2 2 3 4
∂c

Examples
Find thebest −fit values of a∧b so that y =a+bx
x y xy x
2

0 1 0 0
1 1.8 1.8 1
2 3.3 6.6 4
3 4.5 13.5 9
4 6.3 25.2 16
Σ x =10 Σ y=16.9 Σ xy =47.1 2
Σ x =30
Eqns : Σ y=na+ b Σ x , Σ xy =a Σ x+ b Σ x 2

 One Q: Lagrange, Forward difference, Backward difference

 Numerical differentiation
 Simpsons’ 1/3 rule
 Parabola method
Change of Scale
( x−x 0 )
u= , v= y− y 0 , ( his the interval )
h
x−( middle term ) x−( middle of two middle terms )
if n is odd :u= ,if nis even :u=
h 0.5 h
Equations become Σu=na+ b Σv
Example

x−12.5
u= , v= y−20
2.5
x y u v uv u
2

0 12 -5 -8 40 25
5 15 -3 -5 15 9
10 17 -1 -3 3 1
15 22 1 2 2 1
20 24 3 4 12 9
25 30 5 10 50 25
Σu=0 Σ v=0 Σ uv=122 2
Σ u =70
0=6 a+ 0b → a=0122=0 a+70 b → b=1.743→ v=1.743 u

→ ( y−20 ) =1.743
x−12.5
2.5 ( )
→ y−20=0.697 x−8.7142→ y=0.697 x +11.29

x y u=x−1933 v= y−358 uv u
2
u v
2
u
3
u
4

1929 352 -4 -6
1930 356 -3 -2
1931 357 -2 -1
1932 358 -1 0
1933 360 0 2
1934 361 1 3
1935 361 2 3
1936 360 3 2
1937 359 4 1
0 2 51 60 -69 0 708
Σv=na+b Σ u+c Σ u →2=9 a+60 c− [ 1 ] Σ uv=a Σu+ b Σ u + c Σu → 51=60 b → b=0.85
2 2 3

Σ u v=a Σ u +b Σu + c Σ u →−69=60 a+708 c− [ 2 ]

2 2 3 4

20 247
( Eqn 1 )−Eqn2 → =−648 c → c=−0.2673→ a=2.004→ v=2.004 +0.85 u−0.2673 u2
3 3
2
→ y−358=2.004+0.85 ( x−1933 )−0.27 ( x−1933 ) y=−1010135.076+1044.67 x−0.27 x 2
2 2 3 2 2 3 4
Σ y=na+ b Σ x +c Σ x Σ xy =a Σ x+ b Σ x + c Σ x Σ x y=a Σ x + b Σ x +c Σ x