Finite Differences:

1
 Argument & Entry

• Let y f (x) be a function, which assumes the

values f ( a ), f ( a  h ), f ( a  2 h ),......
corresponding to the values of x namely
a , a  h , a  2 h ,........

The each value of x is called argument and

the corresponding values of y is known as
entry.
•

2
 Definition of Forward Difference

• The first forward difference of y f (x) is defined as

f ( x )  f ( x  h)  f ( x )
• The second forward difference of y f (x) is defined as

 f ( x )  f ( x  h)  f ( x )
2

y f (x)
• The general nth forward difference of is defined as
n 1 n 1
 f ( x)  
n
f ( x  h)   f ( x)

Here is the forward difference operator

3
• If the arguments are denoted by x 0 , x 1 , x 2 , .......... . and
the corresponding entries by y 0 , y 1 , y 2 , .......... . then the
first, second, and higher order forward differences are given
by
 y r  y r 1  y r

 2
y r   y r 1   y r
n 1 n 1
 n
y r   y r 1   y r

4
 Forward difference Table
Value of Value of 1st Diff 2nd Diff 3rd Diff 4th Diff 5th Diff
x y
x 0
y 0
 y
x0  h y 1
0
2 y0
 y 1 3 y 0
x0  2h y 2 2 y1
 y 4 y0
2 3 y1
x 0  3h y 3  y2
2 5 y 0
 y 4 y1
3
x0  4h y 3 y 2
4
2 y3
 y 4
x 0  5h y 5

5
• Here y 0 the first entry is called the leading entry and
• y 0 , 2 y 0 , 3 y 0 , 4 y 0 , 5 y 0 are called leading
differences. The quantity h is called the interval 0f
differencing.

6
 Forward Differences in terms of Entries

• Now, 2 y 0   y1   y 0
 2 y 0  ( y 2  y1 )  ( y1  y 0 )
  2 y 0  y 2  2 y1  y 0
• Again,
3 y 0  2 y1  2 y 0
 3 y 0   y 2   y1  ( y1   y 0 )
 3 y 0  ( y 3  y1 )  ( y 2  y1 )  ( y 2  y1 )  ( y1  y 0 )
 3 y 0  y 3  3 y 2  3 y1  y 0 )
Similarly,
4 y0  y4  4y3 6y2  4y1  y0

7
 Problems

I. Construct the forward difference table, given that

x 5 10 15 20 25 30

y 9962 9848 9659 9397 9063 8660

and write down the values of 2 y10 , 3 y 5 , 4 y 5

8
Solution:
x y y  y  y  y  y
2 3 4 5

5 9962
-114
10 9848 -75
-189 2
15 9659 -73 2 -1
-262 -73 1 -1 3
20 9397 -72 2
-334 3
25 9063 -69
-403
30 8660

Now,
 y 10   73
2
 y5  2
3
 y 5  1
4

9
 Problems

1. Construct the forward difference table of the polynomial,

• f ( x)  x3  x 2  2x  1 for values x  0 ( 1 ) 5 .Find the value of
the polynomial at x  6 by extending the table.
• Solution: f ( x)  x3  x 2  2x  1 putting x  0 ,1 , 2 , 3 , 4 , 5
f ( 0 )  1, f (1)  1, f ( 2 )  9 , f ( 3 )  31, f ( 4 )  73 , f ( 5 )  141

• Thus we have,
x 0 1 2 3 4 5

y f (x) 1 1 9 31 73 141

10
• The difference table is

x y y  y
2
 y
3

0 1
0
1 1 8
8 6
2 9 14
22 6
3 31 20
42 6
4 73 26
68 6
5 141 26+6=32
68+32=100 6
6 141+100=241 26+6=32
68+32=100
141+100=241

11
• Now from the above table we have 3 y  6
3
• We know from the higher forward difference
n 1 n 1
 n
y r   y r 1   y r
•   n 1
y   n
y   n 1
y ………………..(i)
r 1 r r

• Now,  y4  y6   y5
 y6   y5  y5
•  y 6  2 y 4   y 4  y 5 (using (i))
 y 6  3 y 3   2 y 3   y 4  y 5
 y 6  6  26  68  141
 y 6  241

12
 Backward Differences

• The forward difference of f (x) is defined as

•
f ( x )  f ( x  h)  f ( x )
Here f (x) is expressed in terms of f (x) and the functional
value one step forward i.e. f (xh)
Now we define backward of f (x) by expressing the difference
using f (x) and the functional value one step backwards.
The backward difference is defined as
f (x)  f (x)  f (x  h)
Here  is called the backward difference operator.

13
• The second forward difference of y f (x) is
defined as
 2 f ( x)  f ( x)  f ( x  h)
• In general
 n f ( x )   n 1 f ( x )   n 1 f ( x  h )

14
 The E -shift Operator
• The shift operator E is defined by
Ef ( x )  f ( x  h )
• Similarly,
E 2 f ( x )  E  Ef ( x )   Ef ( x  h )  f ( x  2 h )
E 3 f ( x )  f ( x  3h )
• In general,
E n
f ( x )  f ( x  nh )
• The operator
E 1
 Ef ( x )  f ( x )
E n
E n

f (x)  f (x)

15
 The operators E and  are commutative

• Here
E  f ( x )   E  f ( x  h )  f ( x ) 
 Ef ( x  h )  Ef ( x )
 f (x  2h)  f (x  h)
 f (x  2h)
   Ef ( x ) 
 E  E

16
 A relation E 1 
• Proof: We have,
f (x)  f (x  h)  f (x)
f (x  h)  f (x)  f (x)
Ef ( x )  (1   ) f ( x )
E  1 

17
 1
A relation   1 E
• Proof: We have,
 f (x)  f (x)  f (x  h)
1
 f (x)  f (x)  E f (x)
1
 f ( x )  (1  E ) f (x)
1
  1 E

18