Scribd
Upload
17 views7 pages

Finite Differences in Numerical Analysis

1. Numerical analysis involves finding numerical solutions to mathematical problems through algorithms and finite differences. 2. The basic approach is to use algorithms, which are well-defined procedures, to obtain numerical answers. 3. Finite differences provide a discrete analog to calculus by examining changes over finite intervals using difference operators like ∆, which yields higher-order differences when applied repeatedly.

Uploaded by

Ruben Reiner Abrencillo
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
17 views7 pages

Finite Differences in Numerical Analysis

1. Numerical analysis involves finding numerical solutions to mathematical problems through algorithms and finite differences. 2. The basic approach is to use algorithms, which are well-defined procedures, to obtain numerical answers. 3. Finite differences provide a discrete analog to calculus by examining changes over finite intervals using difference operators like ∆, which yields higher-order differences when applied repeatedly.

Uploaded by

Ruben Reiner Abrencillo
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Numerical Analysis can be defined as the development and implementation of techniques to find

numerical solutions to mathematical problems.

The basic approach to solve problems in numerical analysis is the Algorithm.

Algorithm is a complete, well-defined procedure for obtaining numerical answer to a given


mathematical problem. It is the essential ingredient of a numerical analysis problem.

Finite Differences
The theory of finite differences is the discrete analog of the derivative.
The name “finite differences” indicates that mathematical analysis proceeds by finite interval rather
than the infinitesimally small intervals in calculus.

The first difference of a function, f (x), denoted found by subtracting two successive functional values
and is defined by
∆ f ( x ) =f ( x+ 1 )−f (x)
The formula assumes that the functional values are tabulated one unit apart.

The more general formula


∆ f ( x ) =f ( x+ h )−f (x )
h
the value of h is called the interval of differencing. Unless otherwise stated, it is assumed that the
interval of differencing is equal to one. It is also assumed that h> 0 and negative intervals of differencing
has no meaning.

The symbol ∆ is not a number but an operator, which when applied to a function changes it into some
other function.

The operator ∆ is often called the forward difference operator or the advancing difference operator

Difference Table
The difference table is the standard format for displaying finite differences.

xo yo
∆ yo
x1 y1
∆ y1
x2 y2
∆ y2
x3 y3
∆ y3
x4 y4
Example: Find ∆ f (x ) if f ( x )=x 2 FOR x={0 ,1 , 2 ,3 , 4 }

Using the formula


∆ f ( x ) =f ( x+ 1 )−f (x)
2 2
∆ f ( 0 )=f ( 0+1 )−f ( 0 )=1 −0 =1
∆ f ( 0 )=f ( 1+1 )−f ( 1 )=22 −12=3
∆ f ( 0 )=f ( 2+ 1 )−f ( 2 ) =32−22=5
∆ f ( 0 )=f ( 3+1 )−f ( 3 )=4 2−32 =7
2 2
∆ f ( 0 )=f ( 4+ 1 )−f ( 4 ) =5 −4 =9

In table
x f (x) ∆ f (x )
0 0
1
1 1
3
2 4
5
3 9
7
4 16

Find the first difference of the function y=2 x−3 for x={0 ,1 , 2 ,3 , 4 } and exhibit the difference table.
∆ f ( x ) =f ( x+ 1 )−f (x)
¿ [ 2 ( x +1 )−3 ] −[ 2 x −3 ]
∆ f ( x ) =2

Thus, ∆ f ( x ) =2 for any value of x .

x f (x) ∆ f (x )
0 −3
2
1 −1
2
2 1
2
3 3
2
4 5

Find the first difference of the function f ( x )=ax


∆ f ( x ) =f ( x+ 1 )−f (x)
∆ ax =( a x+1 )−ax
x x
∆ a =a ( a−1)

Formula: ∆ c y k =c ∆ y k analogous to the differentiation of the function f (cx )


∆ c y k =cy k+1−cy k
¿ c ( y k +1− y k )
∆ c y k =c ∆ y k

Difference formulas
Theorem 4.1 The difference of a constant is zero
∆ c=0

Theorem 4.2 The difference of constant multiplied by a function is


∆ c y k =c ∆ y k

Theorem 4.3 The difference of a sum of two function uk and v k is


∆ ( u k +v k ) =∆ u k + ∆ v k

∆ ( u k +v k ) =(u ¿ ¿ k +1+v k +1)−(u ¿ ¿ k+ v k )¿ ¿


¿ ( u k+1−uk ) +( v k+1−v k )
∆ ( u k +v k ) =∆ u k + ∆ v k

Theorem 4.4 The difference of a product uk v k


∆ (u ¿ ¿ k v k )=u k ∆ v k + v k+1 ∆ uk ¿
∆ (u ¿ ¿ k v k )=u k+1 v k +1−uk v k ¿
¿ uk +1 v k+1−u k v k+1 +uk v k+1−u k v k
¿ v k+1 ( u k+1−u k ) +uk (v k+1 −v k )
∆ (u ¿ ¿ k v k )=u k ∆ v k + v k+1 ∆ uk ¿

Theorem 4.5 The difference of a quotient uk and v k

∆ ( )
uk v k ∆ uk −u k ∆ v k
vk
=
v k +1 v k

∆ ( )
uk u k+1 uk
=
v k v k+1 v k
−

u v −u v
¿ k+1 k k k+ 1
v k+1 v k
u v −u v +u v −u v
¿ k+1 k k k k k k k +1
v k +1 v k
v k ( uk +1−uk ) +u k (v k −v k +1)
¿
v k+1 v k
v k ∆ uk −uk ∆ v k
¿
v k+1 v k
Example
k
Show that if y k =2 , then ∆ y k = y k
k
∆ y k =∆ 2
k+ 1 k
¿ 2 −2
k
¿ 2 ( 2−1 )
k
∆ y k =2
∆ y k= yk

−2 x
Find ∆ y k if y k =3
−2 x
∆ y k =∆ 3
−2 (x+1) −2 x
¿3 −3
−2 x
3 −2 x
¿ −3
9
¿3
−2 x 1
9 ( )
−1
−8 −2 x
∆ y k= ∙3
9

Find ∆ u x v x if u x =2 x and v x =2 x−4


∆ u x v x =u x ∆ v x + v x+1 ∆u x

v x+1=2 ( x +1 )−4
v x+1=2 x−2

∆ u x =2 ( x+1 )−2 x
∆ u x =2

∆ v x =2 ( x +1 )−4−(2 x−4 )
∆ v x =2

∆ u x v x =2 x ( 2 )−( 2 x −2 ) 2
∆ u x v x =8 x −4

By definition
2
∆ u x v x =4 x −8 x
¿¿
∆ u x v x =8 x −4
2 2
If u x =x +1 and v x =−x +1 find
a. ∆ u x v x , and

b. ∆ ( ) ux
vx

Higher Order Differences


The higher order difference can be derived from lower-order differences
∆ 2 f ( x )=∆[∆ f ( x ) ]
¿∆¿¿
¿ ∆ f ( x +1 )−∆ f ( x)
¿ [ f ( x +2 )−f ( x +1 ) ]−[f ( x+ 1 )−f ( x ) ]
2
∆ f ( x ) ¿ f ( x +2 )−2 f ( x +1 ) +f (x)

Note that ∆ 2 does not mean the square of the quantity ∆ , but the process of differencing is to be
performed twice.
∆2 f (x ) ≠ ¿

The third difference of f ( x) is given by the following derivations


∆ 3 f ( x ) =∆2 [∆ f ( x ) ]
¿ ∆ [f ( x+2 )−2 f ( x+1 ) + f ( x ) ]
¿ ∆ f ( x +2 ) −2 ∆ f ( x+ 1 )+ ∆ f ( x ) ¿
¿ [ f ( x +3 )−f ( x+ 2 ) ] −2 [ f ( x+2 ) −f ( x +1 ) ]+[f ( x +1 )−f ( x ) ]
¿ f ( x +3 )−3 f ( x+2 ) +3 f ( x +1)−f (x )

It is important to note that any difference can be expressed in terms of functional values.

The difference of f ( x) are often arranged in tabular format. Such an array is called the difference table

x f (x) ∆ f (x ) ∆ 2 f (x) ∆ 3 f (x) ∆ 4 f (x )


0 f (0)
∆ f (0)
2
1 f (1) ∆ f (0)
∆ f ( 1 ) ∆3 f (0)
2 f ( 2 ) ∆2 f ( 1 ) ∆4 f (0)
3
∆ f ( 2 ) ∆ f (1)
3 f ( 3 ) ∆ 2 f (2)
∆ f (3)
4 f (4)
It should be noted that each column after f (x) is partitioned halfway between the entries in the
preceding column to clearly indicate which two values are subtracted to produce the difference

Each entry in the difference column is the difference of the nearest neighbor to the left.
∆ f ( 0 )=f ( 1 )−f ( 0 )
∆ 2 f ( 0 )=∆[∆ f ( 0 ) ]
¿ ∆ f (1 )−∆ f (0)
∆ f ( 0 )=∆2 f ( 1 )−∆ 2 f (0)
3

Example: verify that the third difference of the third-degree polynomial f ( x )=x 3 are constant
Using the formula ∆ 3 f ( x ) =f ( x+ 3 )−3 f ( x +2 )−3 f ( x +1 )−f (x )
Where f ( x )=x 3
f ( x +1 )=¿ ¿
f ( x +2 )=¿¿
f ( x +3 ) =¿ ¿
∆ f ( x ) =x3 +9 x 2 +27 x+ 27−3 ( x 3 +6 x 2+ 12 x +8 ) +3 ( x 3 +3 x 2+3 x +1 )−x 3
3

3
∆ f ( x ) =6

Difference table of f ( x )=x 3


x f (x) ∆ f (x ) ∆ 2 f (x) ∆ 3 f (x) ∆ 4 f (x )
0 0
1
1 1 6
76
2 8 12
19 6
3 27 18
37
4 64

By constructing a difference table, find the 6th and7th term of the sequence 8, 12, 19, 29, 42
x f (x) ∆ f (x ) ∆ 2 f (x) ∆ 3 f (x) ∆ 4 f (x )

1 8
4
2 123
7
3 19 3
10
4 29 3
13
5 42 3
16
6 58 3
19
7 77

Given the sequence of numbers 1, 8, 27, 64, 125, find n such that (a) ∆ n f (x) is a constant, and
(b) ∆ n f ( x ) =0
x f (x) ∆ f (x ) ∆ 2 f (x) ∆ 3 f (x) ∆ 4 f (x )

1 1
7
2 8 12
19 6
3 27 18 0
37 6
4 64 24
61
5 125

a) n=3
b) n=4

Compute the higher order differences of the following and find n such that ∆ n f ( x ) =0
a. f ( x )=2 x−1
b. f ( x )=x 2
c. f ( x )=x 4

Common questions

Powered by AI

Applying ∆f(x) = f(x+1) - f(x) to a linear function (e.g., y = 2x - 3) yields a constant difference of 2 , indicating uniform growth. For exponential functions such as y_k = 2^k, the difference equals the current function value, highlighting exponential growth rates . This reflects inherent growth patterns—linear functions increase uniformly, while exponential ones grow proportionally, indicating varied implications for modeling and predicting.

Constructing a difference table for f(x) = x^2 involves calculating first and second differences. The first differences are linear, and the second differences are constant and equal, revealing the quadratic growth of the function . This analysis highlights the nature of polynomials—increasing constant higher-order differences align with the polynomial's degree, providing insights into the behavior and structure of the function.

Expressing finite differences in terms of functional values, such as through ∆^n f(x), allows for predictive analysis of sequence behavior. Identifying constant differences in certain degrees indicates the polynomial's degree, while zero differences, as in ∆^n f(x) when n exceeds the function's degree, confirm the polynomial's nature . Such analytical perspectives facilitate understanding function characteristics and potential simplifications for broader math modeling applications.

Higher-order differences such as ∆^2f(x) can be derived from lower-order differences by iteratively applying the difference operator: ∆^2f(x) = ∆[∆f(x)] = f(x+2) - 2f(x+1) + f(x). This derivation is crucial in numerical analysis as it reveals patterns, trends, and stability within data sequences or functions, serving to uncover underlying mathematical properties and forecast behaviors over terms or intervals.

The verification process for the constancy of the third difference of f(x) = x^3 involves computing ∆^3f(x) = f(x+3) - 3f(x+2) + 3f(x+1) - f(x), which simplifies to a constant value of 6 . This constancy is significant as it aligns with the degree of the polynomial; the third differences equal sixth, reflecting the leading coefficient's influence over sequential expansion beyond cubic polynomial terms and highlighting the predictability of polynomial behavior through finite differences.

Theorem 4.4 implies that when calculating the finite difference of a product of two functions, it is essential to incorporate both the finite differences and functional values of each component. Specifically, the formula ∆(u_k v_k) = u_k ∆v_k + v_{k+1}∆u_k resembles the product rule in calculus, demonstrating the complex interdependence and contribution of each function's change to the overall product change . This indicates a deeper understanding of product dynamics as relationships rather than linear combinations.

Applying the first difference operator, ∆f(x) = f(x+1) - f(x), to f(x) = x^2 for x = {0, 1, 2, 3, 4} results in ∆f(0)=1, ∆f(1)=3, ∆f(2)=5, ∆f(3)=7, and ∆f(4)=9 . This reflects a linear increase in the differences, indicating the function's quadratic nature.

The formula ∆u_k v_k = u_k ∆v_k + v_{k+1}∆u_k describes how the finite difference of a product of two functions can be expressed in terms of the finite differences of the individual functions. Specifically, it accounts for the contributions of ∆v_k and ∆u_k by incorporating u_k and v_{k+1}, respectively . This formulation is analogous to the product rule in calculus.

Finite difference operations mirror calculus by translating continuous calculus operations, such as derivation, into discrete analogs. For example, the difference operator ∆ is analogous to a derivative operator. However, they differ in implementation: finite differences deal with incremental discrete changes, suitable for computational and algorithmic solutions . This highlights practical advantages in numerical analysis, particularly in simulations, algorithm design, and discrete data modeling, where exact calculus solutions are inapplicable.

For y_k = 2^k, the first difference ∆y_k = 2^k indicates that the differences between successive terms remain constant and equal to y_k itself, which is a defining characteristic of exponential sequences . This suggests that higher-order differences would remain constant beyond the first difference, showcasing the property that higher-order differences stabilize rapidly for exponential sequences.

You might also like