Finite Differences

Course NO. EEE123

Course Title: Computing Method and Technique for Engineering Analysis
 Finite Differences

• Let y=f(x) be a discrete function.

• If xo, xo+h, xo+2h,…….,xo+nh are the successive values of x, where two consecutive values
differ by a quantity h, then the corresponding values of y are yo, y1, y2,….., yn. The value of
the independent variable x is usually called the argument and the corresponding functional
value is known as the entry. The arguments and entries can be shown in a tabular form as
follows:

Argument xo x1=xo+h x2=xo+2h …….. xn=xo+nh

x

• Entry yo y1 y2 ……… yn
y=f(x) =f(xo) =f(xo+h) =f(xo+2h) =f(xo+nh)

Adharaa Neelim Dewanjee, EEE, CUET 2

 Finite Differences

• To determine the values of f(x) and fʹ(x) etc. for some

intermediate arguments, the following three types of differences
are found useful:

1. Forward differences

2. Backward differences

3. Central differences

Adharaa Neelim Dewanjee, EEE, CUET 3

 Forward Differences

• If we subtract from each value of y (except yo) the preceding value of y, we get y1-yo, y2-
y1,……., yn-yn-1 respectively, known as the first forward differences of y. These results which
may denoted Δyo, Δy1,………. Δyn.
yo  y1  yo , y1  y2  y1 ,........., yn1  yn  yn1 ,
Second Forward
differences. 2 Is called
the second forward
2 yo  (yo )  ( y1  yo )  y1  yo difference operator.
 ( y2  y1 )  ( y1  yo )  y2  2 y1  yo
2 y1  (y1 )  y2  y1  y3  2 y2  y1

.............
2 yn  yn 1  yn  yn  2  2 yn 1  yn

Adharaa Neelim Dewanjee, EEE, CUET 4

 Forward Differences

3 yo  (2 yo )  2 y1  2 yo   (y1 )  (yo ) Third Forward

differences. 3 Is called
 ( y2  y1 )   ( y1  yo )  y2  2y1  yo the second forward
 ( y3  y2 )  2( y2  y1 )  y1  yo difference operator.
 y3  3 y2  3 y1  yo
3 y1  2 y2  2 y1  y4  3 y3  3 y2  y1
.............
3 yn  2 yn 1  2 yn  yn 3  3 yn  2  3 yn 1  yn
• In general the nth differences are defined as

n yk  n 1 yk 1  n 1 yk

Adharaa Neelim Dewanjee, EEE, CUET 5

 Forward Differences

• In function notation, the forward differences are as written below:

f ( x)  f ( x  h)  f ( x)
2 f ( x)  f ( x  2h)  2 f ( x  h)  f ( x)
3 f ( x)  f ( x  3h)  3 f ( x  2h)  3 f ( x  h)  f ( x)

And so on, where h is the interval of differencing.

Adharaa Neelim Dewanjee, EEE, CUET 6

 Forward Differences

• The forward differences are usually arranged in a tabular form in the following manner:
x y=f(x) 1st 2nd 3rd 4th 5th
difference difference difference difference difference

xo yo  f ( xo )
yo
x1  xo  h y1  f ( x1 ) 2 yo
y1 3 yo
x2  xo  2h y2  f ( x2 ) 2 y1 4 yo
y2 3 y1 5 yo
x3  xo  3h y3  f ( x3 ) 2 y2 4 y1
y3 3 y2
x4  xo  4h y4  f ( x4 ) 2 y3
y4
x4  x5  5h y5  Adharaa
f ( x5 ) Neelim Dewanjee, EEE, CUET 7
 Forward Differences

• The first term in the table yo is called the leading term.

• The differences yo , 2 yo , 3 yo ............ are called leading differences.

Adharaa Neelim Dewanjee, EEE, CUET 8

 Forward Differences

• Observation 1:
We can express any higher order forward difference of yo in terms of entries yo, y1, y2,……,yn.
From

yo  y1  yo
2 yo  y2  2 y1  yo
3 yo  y3  3 y2  3 y1  yo
n yo  yn  n C1 yn1  n C2 yn 2 ...... () n yo

Adharaa Neelim Dewanjee, EEE, CUET 9

 Forward Differences

• Observation 2:
We can express any value of y in terms of leading entry yo
From,
y1  yo  yo
y1  yo  yo  (1  ) yo
y2  y1  y1  (1  ) y1  (1  ) 2 yo
y3  (1  )3 yo
yn  (1  ) n yo  yo  n C1yo  n C2 2 yo  ....... n yo

Adharaa Neelim Dewanjee, EEE, CUET 10

 Backward Differences

y1  y1  yo , y2  y2  y1 ,........., yn  yn  yn1 ,

 is the backward difference operator called nabla operator.

 2 y2  (y2 )  ( y2  y1 )  y2  y1

 ( y2  y1 )  ( y1  yo )  y2  2 y1  yo
 2 y3  y3  y2  y3  2 y2  y1

In general,

 n yk   n 1 yk  n 1 yk 1

Adharaa Neelim Dewanjee, EEE, CUET 11

 Backward Differences
• In function notation,
f ( x )  f ( x )  f ( x  h )
f ( x  h )  f ( x  h )  f ( x )
 2 f ( x  2h)  f ( x  2h)  2 f ( x  h)  f ( x )
 3 f ( x  3h)  f ( x  3h)  3 f ( x  2h)  3 f ( x  h)  f ( x)

Adharaa Neelim Dewanjee, EEE, 12

CUET
 Backward Differences

• The backward differences are usually arranged in a tabular form in the following manner:
x y=f(x) 1st 2nd 3rd 4th 5th
difference difference difference difference difference

xo yo  f ( xo )
y1
x1  xo  h y1  f ( x1 )  2 y2
y 2  3 y3
x2  xo  2h y2  f ( x2 )  2 y3  4 y4
 y3  3 y4  5 y5
x3  xo  3h y3  f ( x3 )  2 y4  4 y5
y 4  3 y5
x4  xo  4h y4  f ( x4 )  2 y5
 y5
x4  x5  5h y5  Adharaa
f ( x5 ) Neelim Dewanjee, EEE, CUET 13
 Backward Differences

• Obseration

yn  yn 1  yn
yn 1  yn  yn  (1  ) yn
yn  2  yn 1  yn 1  (1  ) yn 1  (1  ) 2 yn
yn 3  (1  )3 yn
yn  k  (1  ) k yn
yn  k  yn  k C1yn  k C2 2 yn  ......... (1) k  k yn

Adharaa Neelim Dewanjee, EEE, 14

CUET
Adharaa Neelim Dewanjee, EEE, 15
CUET