03 Interpolation

Interpolation
Interpolation
Interpolation is important concept in
numerical analysis.
Quite often functions may not be available
explicitly but only the values of the function
at a set of points.
Interpolation
Interpolation is important concept in
numerical analysis.
Quite often functions may not be available
explicitly but only the values of the function
at a set of points.
The values for f(xi) may be the results from
a physical measurement (conductivity at
different points around UWI)
Interpolation
It may also be from some long numerical
calculation which cant be put into a simple
equation.
Interpolation
It may also be from some long numerical
calculation which cant be put into a simple
equation.
What is required is that we estimate f(x)!
i.e. Draw a smooth curve through xi.
Interpolation
The method of estimating between two
known points (values) is called
interpolation.
While estimating outside of know values is
called extrapolation.
Interpolation
1.
2.
3.
4.
Interpolation is carried out using

approximating functions such as:
Polynomials
Trigonometric functions
Exponential functions
Fourier methods
Interpolation
Theory
Yes approximate but what is a good

approximation?
Clearly a good approximation should be,

such that the error between the true function
and the approximation function should be
very small.

1.
2.
3.
4.
Other than this approximating functions

should have the following properties:
The function should be easy to determine
It should be easy to differentiate
It should be easy to evaluate
It should be easy to integrate
There are numerous theorems on the sorts

of functions, which can be well
approximated by which interpolating
functions.
Generally these functions are of little use.
There are numerous theorems on the sorts

of functions, which can be well
approximated by which interpolating
functions.
Generally these functions are of little use.
The following theorem is useful practically
and theoretically for polynomial
interpolation.
Weierstrass Approximation Theorem

If f(x) is a continuous real-valued function
on [a, b] then for any > 0 , then there
exists a polynomial Pn on [a, b] such that
|(x) Pn(x)| <
for all x [a, b].

This tells us that, any continuous function
on a closed and bounded interval can be
uniformly approximated on that interval by
polynomial to any degree of accuracy.
However there is no guarantee that we will
know f(x) to an accuracy for the theorem to
hold.

Consequently, any continuous function can
be approximated to any accuracy by a
polynomial of high enough degree.
Polynomial Approximation
Polynomials satisfy a uniqueness theorem:
A polynomial of degree n passing exactly
through n + 1 points is unique.
The polynomial through a specific set of points
may take different forms, but all forms are
equivalent. Any form can be manipulated into
another form by simple algebraic rearrangement.
The Taylor series is a polynomial of infinite
order. Thus
(x) = (x0) + '(x0)(x - x0) + 1/2! ''(x0) (x - x0)2+..
However it is impossible computationally to

evaluate an infinite number of terms.
Taylor polynomial of degree n is therefore
usually defined as
(x) = Pn(x) + Rn + 1(x)
where the Taylor polynomial Pn(x) and the
remainder term Rn + 1(x) are given by
Pn(x) = (x0) + '(x0)(x - x0) + + 1/n! n(x0) (x - x0)n
Rn + 1(x) = 1/(n+1)! n+1( ) (x - x0)n+1
where x0<x.
The Taylor polynomial is a truncated Taylor series,
with an explicit remainder, or error term.
The Taylor polynomial cannot be used as an
approximating function for discrete data, because
the derivatives required in the coefficients cannot
be determined.
It does have great significance, however, for
polynomial approximation because it has an
explicit error term.
When a polynomial of degree n, Pn(x), is fitted
exactly to a set of n + 1 discrete data points, (x0, f0),
(x1, f1), , (xn, fn), the polynomial has no error at the
data points themselves. However, at the locations
between the data points, there is an error, which is
defined by
E(x) = (x) - Pn(x)
This error term has the form
E(x) = 1/(n+1)! (x - x0) (x x1) (x xn) n+1( );
x0x.
Interpolation
In Practice
Interpolating Polynomials
Suppose we are given some values, the
principle is that we fit a polynomial curve to
the data.
The reason for this is that polynomials are
well-behaved functions, requiring simple
arithmetic calculations.
Approximating polynomial (interpolating
polynomial) should pass through all the
known points.
Where it does not pass through the points it
should be close to the function.
Approximating
polynomial
(interpolating
polynomial) should
pass through all the
known points.
Where it does not pass
through the points it
should be close to the
function.
True function
Approx 1
Approx 2
Note that the interpolating
polynomial may miss
points of discontinuity.
There is only one
interpolating polynomial
P(xi) or less that matches
the exact values; f(x0),
f(x1),, f(xn) at n+1
distinct base points.
True function
Approx 1
Approx 2
Using Polynomials to approximate a
function given discrete points
We will be looking at two interpolating

methods:
1. Lagrange Interpolation
2. Divided Difference
Lagrange Interpolation
Lagrange Polynomials
A straightforward approach is the use of
Lagrange polynomials.
The Lagrange Polynomial may be used
where the data set is unevenly spaced.
The formula used to interpolate between
data pairs (x0,f(x0)), (x1,f(x1)),, (xn,f(xn)) is
given by,
n
P x Pj x
j 1
Where the polynomial Pj(x) is given by,

n
Pj x y j
k 1
k j
x xk
x j xk
In general,
x x2 x x3 ... x xn
P x y1
x1 x2 x1 x3 ... x1 xn
x x1 x x3 ... x xn
y2
...
x2 x1 x2 x3 ... x2 xn
x x1 x x2 ... x xn 1
yn
xn x1 xn x2 ... xn xn 1
Consider the table of interpolating points
we wish to fit.
i
f(x)
x0
f(x0)
x1
f(x1)
x2
f(x2)
x3
f(x3)
i
f(x)
x0
f(x0)
x1
f(x1)
x2
f(x2)
x3
f(x3)
The interpolation polynomial is,
x x1 x x2 x x3
x x0 x x2 x x3
P x
f x0
f x1
x0 x1 x0 x2 x0 x3
x1 x0 x1 x2 x1 x3
x x0 x x1 x x3 f x x x0 x x1 x x2 f x
x2 x0 x2 x1 x2 x3 2 x3 x0 x3 x1 x3 x2 3
Note that the Lagrangian polynomial passes
through each of the points used in its
construction.
Advantages
The Lagrange formula is popular because it
is well known and is easy to code.
Also, the data are not required to be
specified with x in ascending or descending
order.
Disadvantages
Although the computation of Pn(x) is simple, the
method is still not particularly efficient for large
values of n.
When n is large and the data for x is ordered, some
improvement in efficiency can be obtained by
considering only the data pairs in the vicinity of the
x value for which Pn(x) is sought.
The price of this improved efficiency is the
possibility of a poorer approximation to Pn(x).
Diagram showing Interpolation

(incrementally from one to 5 points)
Newtons Divided differences

The nth degree polynomial may be written
in the special form:

The nth degree polynomial may be written
in the special form:
If we take ai such that Pn(x) = (x) at n+1
known points so that Pn(xi) = (xi), i=0,1,
,n, then Pn(x) is an interpolating
polynomial.

A divided difference is defined as the
difference in the function values at two
points, divided by the difference in the
values of the corresponding independent
variable.
Thus, the first divided difference at point is
defined as
f 0 f1
f x0 , x1
x0 x1

Thus, the first divided difference at point is
f 0 f1
defined as f x0 , x1 x x
0
1
The second difference is given as:
f x0 , x1 f x1 , x2
f x0 , x1 , x2
x0 x2
In general,
f x0 ,..., xn 1 f x1 , xn
f x0 , x1 ,..., xn
x0 xn

A divided difference table.

One with actual values.
The 3rd degree polynomial fitting all points from x0 = 3.2

to x3 = 4.8 is given by
P3(x) = 22.0 + 8.400(x - 3.2) + 2.856(x - 3.2)(x - 2.7)
0.528(x - 3.2)(x - 2.7)(x - 1.0)
The 4th degree polynomial fitting all points is given by
P4(x) = P3(x) + 0.256(x - 3.2)(x - 2.7)(x - 1.0)(x - 4.8)
The interpolated value at x = 3.0 gives P3(x) = 20.2120.
There are two disadvantages to using the

Lagrangian interpolation polynomial for
interpolation.
1. It involves more arithmetic operations than does
the divided differences.
2. If we desire to add or subtract a point from the
set to construct the polynomial, we essentially
have to start over in the computations.
The divided difference avoids this.

Tabular data have a finite number of digits.
The last digit is typically rounded off.
Round off has an effect on the accuracy of
the higher-order differences.
A Brief Word on Fitting Data

Consider the table of data.
i
f(x)
x0
f(x0)
x1
f(x1)
x2
f(x2)
x3
f(x3)
Assume that in small regions the data can

be approximated by a polynomial of low
degree.
P x a0 a1 x a2 x 2 ... an 1 x x 1

Because the fit is local there is a different
polynomial for each region of the table.
If the degree of the polynomial is low, many
polynomials are needed to fit the entire
region.
These fits may behave better than one
higher degree polynomial.
Fit using an eighth-degree Fit using a series of 3rd

polynomial
degree polynomials
True Curve

Although it is tempting, higher order
polynomials should not be used unless there
is reason to believe that using one
polynomial will give a good fit.

03 Interpolation

Uploaded by

Copyright:

Available Formats

03 Interpolation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

03 Interpolation

Uploaded by

Copyright:

Available Formats

Interpolation

Interpolation is carried out using

Yes approximate but what is a good

Clearly a good approximation should be,

Other than this approximating functions

There are numerous theorems on the sorts

There are numerous theorems on the sorts

Weierstrass Approximation Theorem

Weierstrass Approximation Theorem

Weierstrass Approximation Theorem

Weierstrass Approximation Theorem

However it is impossible computationally to

Rn + 1(x) = 1/(n+1)! n+1( ) (x - x0)n+1

We will be looking at two interpolating

Where the polynomial Pj(x) is given by,

The interpolation polynomial is,

Diagram showing Interpolation

Newtons Divided differences

Newtons Divided differences

Newtons Divided differences

Newtons Divided differences

Newtons Divided differences

Newtons Divided differences

Newtons Divided differences

Newtons Divided differences

The 3rd degree polynomial fitting all points from x0 = 3.2

Newtons Divided differences

There are two disadvantages to using the

Newtons Divided differences

A Brief Word on Fitting Data

Assume that in small regions the data can

A Brief Word on Fitting Data

Fit using an eighth-degree Fit using a series of 3rd

A Brief Word on Fitting Data

You might also like