0% found this document useful (0 votes)
64 views4 pages

Lagrange's Interpolation Numerical

1) The document describes Gauss's backward interpolation formula for constructing a polynomial that interpolates given data points. 2) The formula uses 2n distinct interpolating points and constructs a polynomial of degree at most 2n-1 that passes through each point. 3) The coefficients of the polynomial are determined explicitly in terms of the points and function values at the points, with the leading coefficient involving a product of differences between the points.

Uploaded by

Rinka
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
64 views4 pages

Lagrange's Interpolation Numerical

1) The document describes Gauss's backward interpolation formula for constructing a polynomial that interpolates given data points. 2) The formula uses 2n distinct interpolating points and constructs a polynomial of degree at most 2n-1 that passes through each point. 3) The coefficients of the polynomial are determined explicitly in terms of the points and function values at the points, with the leading coefficient involving a product of differences between the points.

Uploaded by

Rinka
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

INTERPOLATION 35

(d) Gauss's backward formula with


Let flx) be known for
even
interpolation points
2n distinct
x X t ih, i
=
1, 2, = interpolating points xg» *+ X+ 2
yt =flx,), i 0, 1, Xln-11, X such that
..., n -1,
xh and *- =
0
Let p(x) be the =

interpolating polynomial of degree (2n - 1) at most y_, =flx_). n -


..

yi = 0, tl, t2,... + (n - 1) and p(x,) =fx_,) y- such that


p(x,) flx) =
= =

Let plx) Bo+ B,(x X) B,x- x.) (x


=
-

B2-2x-*-{n=1) (* -X-(n-2)) (x x-1) (x x) B,(x


-

+
x_)
x1) (x - x)
x) (x
(x - x)
x,)
-

-
- +.
+
+
-
-

x) (x x1).. (x
+B2n-1X-*-(n-1)) (xX X-(n-2)) (x x) (x - x) (x -
-X-2) -
-

The coefficient's
B,s x). (x
are (x -Xn-) -

X-1)
Bo Yo
2r-
B2r-1 r = 1,2, n
(2r-1)!H7T=l, 2, ., n
B2 r-1)
(2r)!2
=s =
1, 2,.
2, . . , n -
l

plx)= o+Ay+ A+u 3! ) ,ya


+.+ u(u-)...(u-(n-2))(u-(n-))n-2
(2n-2)! Y-n-1)
u-1)...(u-(n-2)(u-
(2n-1)!
(n-1) "y-n
where u =
Error term
Here the error term
R2,(x) is similar as
given in Gauss's forward
interpolating points interpolation with odd

i.e. R2,) u(u' 12) (u2


= -
-

22) ..
(u -

(n 1)) (u n)
-

+
where min{xo *^1» X(n-1)» X X} , < < (2n)!
(e) Gauss's third interpolation formula
max{xo X1 X(n-1) X- x}

Here we starts with x, instated of xg and runs parallel to the backward formula. Thus
coefficients
Ay A'y - and we set u
are !h' 21h 3!h4!h =
(i.e. u
replace by u -

1). These changes


reduce the backward formula to

P(x) =

y + A+M 2!
y+ U-D(u-2)
3!

ulu-1)(u-2ay+
4!
-1)(u-2)(u-3)5,
5!
uu-1)(u -2)...(u-(n-2) Mu-(n-1))%u-n)
(21-1)!
Here the number of interpolating points are even 1.e.
Xo» X1 *2., z(n-1) n

w
IN C PROGRAMMING
6 NUMERICAL METHODS WITH BASIC CONCEPTS

Error term:
R2,(x) = u(u2 - 12) (u2 - 22).. (u2 - (n - 1)) (u - n) (2m)!

C n"o» X^19 .Xn-1)» x x} <g < max{x, X1 ¥a(n-1)» *r* *}

NOTE 1 : Gauss's forward interpolation formula is used for interpolating at the startina
of the
centre of the interpolating points with 0 < u < l to mininmize the error.
NOTE 2 Gauss's backward interpolation formula is used for interpolating at the ending ofs
i the
centre of the interpolating points with -1 < u< 0 to minimize the error.

3.10 STIRLING's CENTRAL INTERPOLATION FORMULA


Taking the mean of Gauss's forward and backward formula with odd interpolating points
,
X22 *n We obtain
Px) = Y+
Px) = y, u+Ay-1 , u(u+)+uu-) ^2y.+ u{u-1)
2 2!2 3!
Ay+2++
y+Ay
3
uu-1)(u-2)..(u-n-1) Ayn-)+Ay_
(21-1)! 2

u(u-1)...(u-(n-1))(u-n)+u (u*-1)..(uf -(n-1*) (u +n)


2.(2m)!

-)y+lue
4! e-P-
5!
-2)A',+a +

+
u(u-1) (" -2)... (u- (n-1)) A"y(n=1)
(2n-1)!
+Ay-
u (u-1)(u-2*)..(uf -(n-1)
(2n)! y
where the error term

R2n+1) = (u -1)u-2*)..(uf -n*)


(2n+1)! f**"E)+f
2
6
where min{xo. *^1 X} < <
K S1 `2 max\x» * x,w x}
Doins
NOTE: Stirling's formula is used for the
with to
interpolation of the central of the interpolans
u minimize the error.
3.11 BESSEL'S CENTRAL INTERPOLATION FORMULA
Ven number «
Taking the mean of Gauss's forward and
:
third
interpolating points *o» *21 *2 We
interpolation formula with eve
*2{n-1» X> obtain
4-
pt)- + ) &,+ay.u-)utu-1)
3!
4-uu-1)..(u-(n-2*
(2n-1)!
(u-(n-1) A yn-l)
PROGRAMMINGG
CONCEPIS IN C
WITH BASIC
NUMERICAL METHODS
formula
Difference table for Bessel's

y Ay
Ay
Y4

y4
y
y
y.3
y- 8y
By1

y
Dy

3.12 LaGRANGE'S INTERPOLATION FORMULA


Let flx) be known at (n + 1) interpolating points xo X .., X, and y; =flx), i = 0, 1, 2,
where x- Xo X2- X *, - Xn-1 need not necessary be equal.
Let L,(x) be the interpolating polynomial of
i = 0, 1, 2, .., n.
n at most degree such that L,(x) =J4
Since L,(x) is a
polynomial of degree n at most, it
may be written as

L,) Agx x) x - *2). (*x)+ A,(x x) (x -x). (x -x,


= -

Ax- o) * *) x x- x)+ + A(x x) (x x). (x Xa-


- -

-
-
-

+
A,x ) (x- x1) (x-n-
-
-

(1)
Substitutingx = x, in (1), we obtained

i(x,-X,)x, -z,).. x, -xx-X T 0, 1, 2 "


Equation (1) can be written as

L) A,(x,-x,)x-x)(x-X,)x-x,,).. (x-x,)
1=0
INTERPOLATION 39

o)X-*) (X-X,)(x-X,,)...(,x-x,)
X, *)(a,-x,)... (x, - x,-)(x-x)...(,x-x,)

=0

(x-x)x-x)..
(x,-x)(x -x,)...(x,,xX,-)*-Xi)
where ) is known as
Lagrangien function.
-xj-)(x, -X;1)4, *,)
Let wlx) =
(x -

x) (x -x) ...
(x -

x,)
:. Wx) =
(x -

x) (x -x). (x -

x,) + (x -

x) (x x) -
. .
(x x,)
-

+ X -x) (x -x).. (x -x-

Hence x) =
w(x)
(x-x,)wx,)
L=t-)waj 2
Error term

The error term is R(x) =


(x - *) (x - x).. (x -

x,) (n+1)!
=

n+1)!
where min{xg, X ,
X Xo» X1 ..n X}
X} << max

where min{xo» X ,
X max{*» X1 Xn X}
X}< <
interpolation formula becomes
NOTE 1: For two interpolating points xo X the Lagrenge's

L,x)= 1
L) y,+
linear interpolation formula
This formula is known as

is unique.
NOTE 2 Lagrange's interpolation polynomial of degree at most n, such that
distinct from L,(x)
Let L,(x) be another polynomial,
i =0, 1, ..
L,x) =fa) =y,
Let ylx) L,(x) - L, (x) =

0, i =
0, 1, 2, n.
-y;
,
=

y(x) L,(x) L,(a) yi


=
-

at most with at
least (n+1) zeros.

polynomial of degree n
Thusy(x) is a
Hence y(x) = 0

L(x)= L,()
Thus L,(x) is unique.
FUNCTIONS
LAGRANGIAN
PROPERTIES OF
3.13 linear
transformation.
invarient under
functions remain
(a) Lagrangian 0.
Let x = a + Bt where ß *

Then x = a + Bt
=B(;
-

x- B(t -1)and x;-

You might also like