Polynomial Interpolation of Functions: Proofs and Formulas - Prof. Raytcho Lazarov, Assignments of Mathematical Methods for Numerical Analysis and Optimization

The homework problems for math 609-602, focusing on polynomial interpolation of functions of one variable. Topics include proving properties of divided differences, leibniz formula, finding interpolating polynomials, and estimating interpolation errors.

Typology: Assignments

Pre 2010

Uploaded on 02/10/2009

koofers-user-4b7
koofers-user-4b7 🇺🇸

10 documents

1 / 1

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
MATH 609-602
Homework #5
Polynomial interpolation of functions of one variable
(1) Let f(x) = xnand f[x0, x1, ..., xn] be the divided difference of order nusing
the points x0< x1< ... < xn. Prove that:
(a) (10 pts) f[x0, x1, ..., xn] = 1;
(b) (10 pts) f[x0, x1, ..., xn1] = x0+x1+... +xn1.
(2) (10 pts) If f[x0, x1, . . . , xn] denotes the divided difference of order nprove the
Leibnitz formula
(fg)[x0, x1, . . . , xn] =
n
X
k=0
f[x0, x1, . . . , xk]g[xk, xk+1, . . . , xn].
(3) (10 pts) Find the Lagrange and backward Newton divided difference interpo-
lating polynomials for the data (0,1),(0.5,2),(1,3),(1.5,4).
(4) (20 pts) Let
Ln,k(x) = (xx0)...(xxk1)(xxk+1)...(xxn)
(xkx0)...(xkxk1)(xkxk+1)...(xkxn).
Show that for any xthe following relations are valid:
(a) Pn
k=0 Ln,k(x) = 1;
(b) Pn
k=0 xm
kLn,k(x) = xmfor m= 1, ..., n.
(5) (20 pts) Estimate the interpolation error of cos xin the interval (0,0.4) by a
polynomial of degree 2 using the interpolation nodes x0= 0, x1= 0.2, x2=
0.4.
(6) (20 pts) Pn
k=0(xxk)mLn,k (x) = 0 for m= 1, ..., n.
In your free time and for your amusement:
Show that if ω(x) = (xx0)(xx1)...(xxn) then:
(1) Pn
k=0(xxk)n+1Ln,k (x) = (1)nω(x)
(2) Pn
k=0(xxk)n+2Ln,k (x) = (1)nω(x)Pn
k=0(xxk)
(3) Pn
k=0 Ln,k(0)xn+1
k= (1)nx0.x1...xn.
1

Partial preview of the text

Download Polynomial Interpolation of Functions: Proofs and Formulas - Prof. Raytcho Lazarov and more Assignments Mathematical Methods for Numerical Analysis and Optimization in PDF only on Docsity!

MATH 609-

Homework # Polynomial interpolation of functions of one variable

(1) Let f (x) = xn^ and f [x 0 , x 1 , ..., xn] be the divided difference of order n using the points x 0 < x 1 < ... < xn. Prove that: (a) (10 pts) f [x 0 , x 1 , ..., xn] = 1; (b) (10 pts) f [x 0 , x 1 , ..., xn− 1 ] = x 0 + x 1 + ... + xn− 1.

(2) (10 pts) If f [x 0 , x 1 ,... , xn] denotes the divided difference of order n prove the Leibnitz formula

(f g)[x 0 , x 1 ,... , xn] =

∑^ n

k=

f [x 0 , x 1 ,... , xk]g[xk, xk+1,... , xn].

(3) (10 pts) Find the Lagrange and backward Newton divided difference interpo- lating polynomials for the data (0, 1), (0. 5 , 2), (1, 3), (1. 5 , 4).

(4) (20 pts) Let

Ln,k(x) =

(x − x 0 )...(x − xk− 1 )(x − xk+1)...(x − xn) (xk − x 0 )...(xk − xk− 1 )(xk − xk+1)...(xk − xn)

Show that for any x the following relations are valid: (a)

∑n k=0 Ln,k(x) = 1; (b)

∑n k=0 x m k Ln,k(x) =^ x m (^) for m = 1, ..., n.

(5) (20 pts) Estimate the interpolation error of cos x in the interval (0, 0 .4) by a polynomial of degree 2 using the interpolation nodes x 0 = 0, x 1 = 0. 2 , x 2 = 0 .4. (6) (20 pts)

∑n k=0(x^ −^ xk) mLn,k(x) = 0 for m = 1, ..., n.

In your free time and for your amusement:

Show that if ω(x) = (x − x 0 )(x − x 1 )...(x − xn) then:

(1)

∑n k=0(x^ −^ xk) n+1Ln,k(x) = (−1)nω(x) (2)

∑n k=0(x^ −^ xk) n+2Ln,k(x) = (−1)nω(x) ∑n k=0(x^ −^ xk) (3)

∑n k=0 Ln,k(0)x

n+ k = (−1)

nx 0 .x 1 ...xn.

1