hw10_solutions - Homework 10 Math 471, Fall 2007 Assigned:...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Homework 10 Math 471, Fall 2007 Assigned: Friday, November 16, 2007 Due: Monday, December 3, 2007 Include a cover page Clearly label all plots using title , xlabel , ylabel , legend Use the subplot command to compare multiple plots Include printouts of all Matlab code, labeled with your name, date, section, etc. (1) (Piecewise Linear Interpolation) Use the error theorem on p. 390 to compute an upper bound on the pointwise error from interpolating f ( x ) = e- x 2 on the interval [- 4 , 4] with piecewise linear functions at the points- 4,- 2, 0, 2, and 4. max x [- 4 , 4] | e- x 2- s ( x ) | 1 8 h 2 max x [- 4 , 4] | f 00 ( x ) | = 1 . (2) (Cubic Spline Interpolation) P. 403 #10, P. 404 #15 P. 403 #10a j a j b j c j d j 0 0.500 1.49 0.286-0.153 1 1.43 2.18 0.579-0.153 2 2.64 2.64 0.350-0.330 3 4.01 2.75 -0.144 -0.330 P. 403 #10b j a j b j c j d j 0 0.500 1.50 0.756-0.106 1 1.43 2.18 0.600-0.175 2 2.64 2.64 0.334-0.287 3 4.01 2.76 -0.0961 -0.478 P. 404 #15 j a j b j c j d j 0 0.500 1.72 0.523 1 1.43 2.11 0.784-0.297 2 2.64 2.67 0.338-0.426 3 4.01 2.69 -0.301 0.200 (3) (Richardson Extrapolation) P. 454 #11 (a) Adding the Taylor series for f ( x + h ) and f ( x- h ) gives f ( x + h ) + f ( x- h ) = 2 f ( x ) + h 2 f 00 ( x ) + h 4 12 f (4) ( x ) + h 8 2880 f (8) ( x ) + h 8 2880 [ f (8) ( )- f (8) ( x )] . Thus 1 2 f 00 ( x ) = f ( x + h )- 2 f ( x ) + f ( x- h ) h 2- h 2 12 f (4) ( x )- h 4 360 f (6) ( x )- h 6 2880 f (8) ( x ) + o ( h 6 ) . (b) h O ( h 2 ) O ( h 4 ) O ( h 6 ) 0.5 2.255252 0.25 2.062826 2.049998 0.125 2.015645 2.012500 2.010000 (4) (NewtonCotes). Suppose that f is a function with four continuous derivatives on the interval [ a,b ]. Recall that the error bound for the composite trapezoidal rule T ( h ) with panel width h is T ( h )- R b a f ( x )d x = ( b- a ) h 2 12 f 00 ( ) for some [ a,b ]. The error in the composite Simpsons rule S ( h ) with panel width h is S ( h )- R b a f ( x )d x = ( b- a ) h 4 180 f (4) ( ) for some (different) [ a,b ]. (a) Let f ( x ) = e- x sin x . For the composite trapezoidal rule and the composite Simpsons rule, find the number of panels n required to integrate f on the interval [0 , 2 ] with error at most 10- 4 . Recall that h = 2 /n . How many function evaluations are required in each case? The first five derivatives of f are f ( x ) = e- x (cos x- sin x ) f 00 ( x ) =- 2e- x cos x f 000 ( x ) = 2e- x (sin x + cos x ) f (4) ( x ) =- 4e- x sin x f (5) ( x ) = 4e- x (sin x- cos x ) To develop the bound for the composite trapezoidal rule, we need to calculate the maximum value of | f 00 ( x ) | on the interval [0 , 2 ]. The extremum must occur at]....
View Full Document

This note was uploaded on 04/02/2008 for the course MATH 471 taught by Professor Linzhi during the Winter '08 term at University of Michigan.

Page1 / 7

hw10_solutions - Homework 10 Math 471, Fall 2007 Assigned:...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online