NUMERICAL DIFFERENTIATION

The derivative of f (x) at x0 is:

f x0

Limit
h 0

f x0 h f ( x0 )
h

An approximation to this is:

f x0 h f ( x0 )
f x 0
h

for small values of h.

Forward Difference
Formula

Let f ( x ) ln x and x0 1.8

Find an approximate value for f 1.8

f (1.8)

f (1.8 h)

0.1 0.5877867 0.6418539

f (1.8 h) f (1.8)
h
0.5406720

0.01 0.5877867 0.5933268

0.5540100

0.001 0.5877867 0.5883421

0.5554000

The exact value of

f 1.8 0.55 5

Assume that a function goes through three points:

x0 , f x0 , x1 , f x1 and x2 , f x2 .
f ( x) P( x)
P x L0 x f x0 L1 x f x1 L2 x f x2
Lagrange Interpolating Polynomial

P x L0 x f x0 L1 x f x1 L2 x f x 2
( x x1 )( x x2 )
P x
f x0
( x0 x1 )( x0 x2 )
( x x0 )( x x2 )

f x1
( x1 x0 )( x1 x2 )
( x x0 )( x x1 )

f x2
( x2 x0 )( x2 x1 )

f ( x ) P ( x )

2 x x1 x2
P x
f x0
( x0 x1 )( x0 x2 )
2 x x0 x2

f x1
( x1 x0 )( x1 x2 )
2 x x0 x1

f x2
( x2 x0 )( x2 x1 )

If the points are equally spaced, i.e.,

x1 x0 h and x 2 x0 2h
2 x0 ( x0 h) ( x0 2h)
P x0
f x0
x0 ( x0 h) x0 ( x0 2h)
2 x0 x0 ( x0 2h)

f x1
( x0 h) x0 ( x0 h) ( x0 2h)
2 x0 x0 ( x0 h)

f x2
( x0 2h) x0 ( x0 2h) ( x0 h)

 3h
2h
h
P x 0
f x0
f x1 2 f x 2
2
2
2h
h
2h
1
P x0
3 f x0 4 f x1 f x2
2h
Three-point formula:

1
3 f x0 4 f x0 h f x0 2h
f x0
2h

If the points are equally spaced with x0 in the middle:

x1 x0 h and x 2 x0 h

2 x0 ( x0 h) ( x0 h)
P x0
f x0
x0 ( x0 h) ( x0 ( x0 h)
2 x0 x0 ( x0 h)

f x1
( x0 h) x0 ( x0 h) ( x0 h)
2 x0 x0 ( x0 h)

f x2
( x0 h) x0 ( x0 h) ( x0 h)

0
h
h
P x 0
f x0 2 f x1 2 f x 2
2
h
2h
2h
Another Three-point formula:

1
f x0 h f x0 h
f x0
2h

Alternate approach (Error estimate)

Take Taylor series expansion of f(x+h) about x:

h2 2
h3 3
f x h f x hf x
f x
f x
2
3!
h2 2
h3 3
f x h f x hf x
f x
f x
2
3!
f x h f x
h 2
h2 3
f x f x
f x
h
2
3!
.............. (1)

f x h f x
f x O h
h
f x h f x
f x
O h
h
f x h f x
f x
h

Forward Difference
Formula

h 2
h2 3
O h f x
f x
2
3!

4h 2 2
8h 3 3
f x 2h f x 2hf x
f x
f x
2
3!

4h 2 2
8h 3 3
f x 2h f x 2hf x
f x
f x
2
3!
f x 2h f x
2h 2
4h 2 3
f x
f x
f x
2h
2
3!
................. (2)

f x h f x
h 2
h2 3
f x f x
f x
h
2
3!
.............. (1)
f x 2h f x
2h 2
4h 2 3
f x
f x
f x
2h
2
3!
................. (2)

2 X Eqn. (1) Eqn. (2)

f x h f x f x 2 h f x
2

h
2h
2
3
2h 3
6h 4

f x
f x
f x
3!
4!
f x 2 h 4 f x h 3 f x
2h
2 h2 3
6 h3 4
f x
f x
f x
3!
4!
f x O h2

 f x 2h 4 f x h 3 f x
f x O h2
2h

f x 2h 4 f x h 3 f x
f x
O h2
2h

f x 2h 4 f x h 3 f x
f x
2h
Three-point Formula

2
3
2
h
6
h
O h2
f 3 x
f 4 x
3!
4!

The Second Three-point Formula

Take Taylor series expansion of f(x+h) about x:

h2 2
h3 3
f x h f x hf x
f x
f x
2
3!
Take Taylor series expansion of f(x-h) about x:

h2 2
h3 3
f x h f x hf x
f x
f x
2
3!
Subtract one expression from another
2h 3 3
2h 6 6
f x h f x h 2hf x
f x
f x
3!
6!

2h 3 3
2h 6 6
f x h f x h 2hf x
f x
f x
3!
6!

f x h f x h
h2 3
h5 6
f x
f x
f x
2h
3!
6!
f x h f x h h 2 3
h5 6
f x

f x
f x
2h
3!
6!

f x h f x h
f x
O h2
2h

2
5
h
h
O h2
f 3 x
f 6 x
3!
6!

f x h f x h
f x
2h
Second Three-point Formula

Summary of Errors
f x h f x
f x
h

Error term

Forward Difference
Formula

h 2
h2 3
O h f x
f x
2
3!

Summary of Errors continued

First Three-point Formula
f x 2h 4 f x h 3 f x
f x
2h

2
3
2
h
6
h
3
4
2

x
O
h

f
x

f
Error term
3!
4!

Summary of Errors continued

Second Three-point Formula

f x h f x h
f x
2h

2
5
h
h
3
6
2

x
O
h

f
x

f
Error term
3!
6!

Example:

f x xe

Find the approximate value of

f x

1.9

12.703199

2.0

14.778112

2.1
2.2

17.148957
19.855030

f 2

with

h 0.1

Using the Forward Difference formula:

1
f x0 f x0 h f x0
h

1
f 2
f 2.1 f 2

0.1
1

17.148957 14.778112
0.1
23.708450

Using the 1st Three-point formula:

1
3 f x0 4 f x0 h f x0 2h
f x0
2h

1
3 f ( 2) 4 f ( 2.1) f ( 2.2)
f 2
2 0.1
1
3 14.778112 4 17.148957

0.2
19.855030
22.032310

Using the 2nd Three-point formula:

1
f x0 h f x0 h
f x0
2h

1
f ( 2.1) f (1.9)
f 2
2 0.1
1
17.148957 12.703199

0.2
22.228790
The exact value of

f 2 is : 22.167168

Comparison of the results with h = 0.1

The exact value of f 2 is 22.167168
Formula

f 2

Error

Forward Difference

23.708450

1.541282

1st Three-point

22.032310

0.134858

2nd Three-point

22.228790

0.061622

Second-order Derivative
h2 2
h3 3
f x h f x hf x
f x
f x
2
3!
h2 2
h3 3
f x h f x hf x
f x
f x
2
3!

Add these two equations.

2h 2 2
2h4 4
f x h f x h 2 f x
f x
f x
2
4!
2h2 2
2 h4 4
f x h 2 f x f x h
f x
f x L
2
4!

2
f x h 2 f x f x h
2
h
2
4

f
x

f
2
h
4!
2

f
x

2
f
x

f
x

h
2
h
2
4

f x

f x
2
h
4!

f x h 2 f x f x h
x
h2

NUMERICAL INTEGRATION
b

f ( x )dx
a

area under the curve f(x) between

x a to x b.

In many cases a mathematical expression for f(x) is

unknown and in some cases even if f(x) is known its
complex form makes it difficult to perform the integration.

y
f(b)

f(x)

f(a)

x 0 =a

x 0 =b

f(b)
f(x)
f(a)

x 0 =a

x 0 =b

Area of the trapezoid

The length of the two parallel sides of the trapezoid
are: f(a) and f(b)
The height is b-a
b

ba
f ( a ) f ( b )
f ( x )dx
2
h
f ( a ) f ( b )
2

Simpsons Rule:

x2

x0

x2

f ( x )dx P ( x )dx
x0

x1 x0 h and x 2 x0 2h

x2
x0

P x dx

x2
x0

x2

x0

x2
x0

( x x1 )( x x2 )
f x0 dx
( x0 x1 )( x0 x2 )

( x x0 )( x x2 )
f x1 dx
( x1 x0 )( x1 x2 )

( x x0 )( x x1 )
f x2 dx
( x2 x0 )( x2 x1 )

x2

x0

x2

f ( x )dx P ( x )dx
x0

h
f x0 4 f ( x1 ) f ( x 2 )
3

f(b)
f(x)
f(a)

x 0 =a

x 0 =b

f(b)
f(x)
f(a)

x 0 =a

x 0 =b

f(b)
f(x)
f(a)

x 0 =a

x 0 =b

f(b)
f(x)
f(a)

x 0 =a

x 0 =b

Composite Numerical Integration

Riemann Sum
The area under the curve is subdivided into n
subintervals. Each subinterval is treated as a
rectangle. The area of all subintervals are added
to determine the area under the curve.
There are several variations of Riemann sum as
applied to composite integration.

 xIn Left
a / n
b Riemann
x1 sum,
a the leftx2
x3

side sample of

x is
theafunction
used as the
height
a 2ofthe
x
individual
M
rectangle.

xi a i 1 x

b
a

f x dx f xi x
i 1

x b a / n

In Right Riemann
sum,
x1 the
a right-side
x
sample of the
x2 a is2used
x as
function
the height of the
x3 a 3rectangle.
x
individual

M
xi a i x
n

f x dx f x x
b

i 1

 x In bthe
aMidpoint
/n
Rule, the sample at
x1 the
a middle
2 1 1of
x / 2
the
x2 subinterval
a 2 2 1is
x / 2
used
as the height of the
x3 individual
a 2 3 1 x / 2
Mrectangle.
xi a 2 i 1 x / 2
n

f x dx f x x
b

i 1

Composite Trapezoidal Rule:

Divide the interval into n subintervals and apply
Trapezoidal Rule in each subinterval.
b

f ( x )dx f a 2 f ( x k ) f (b )
2
k 1

n 1

where

ba
h
and x k a kh for k 0, 1, 2, ... , n
n

Find

sin (x)dx
0

by dividing the interval into 20 subintervals.

n 20
ba
h

n
20
k
x k a kh
, k 0, 1, 2, ....., 20
20

h
f ( x k ) f ( b )
0 sin( x )dx 2 f a 2
k 1

n 1

19

sin 0 2 sin
sin

40
20
k 1

1.995886

Composite Simpsons Rule:

Divide the interval into n subintervals and apply
Simpsons Rule on each consecutive pair of subinterval.
Note that n must be even.

h
f ( x )dx f a 2
3

( n / 2 ) 1

f (x
k 1

2k

n/ 2

k 1

4 f ( x 2 k 1 ) f (b)

where

ba
h
and x k a kh for k 0, 1, 2, ... , n
n

Find

sin (x)dx
0

by dividing the interval into 20 subintervals.

ba
n 20
h

n
20
k
x k a kh
, k 0, 1, 2, ....., 20
20

9

2k
sin

0 sin( x )dx 60 sin 0 2

20
k 1

( 2k 1)
4 sin
sin( )
20

k 1

10

2.000006