Final Exam on Numerical Computations | MATH 451, Exams of Mathematics

Material Type: Exam; Class: Numerical Computations; Subject: Mathematics; University: Penn State - Main Campus; Term: Fall 1998;

Typology: Exams

Pre 2010

Uploaded on 09/24/2009

koofers-user-suj
koofers-user-suj 🇺🇸

10 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Next: About this document ...
CSE/MATH 451 Final, Fall 1997
Name: I.D. Number:
1 2 3 4 5 total
1. (25 points) Let N be a given positive number.
(1) Use Netwon’s method on the equation x2-N=0to derive the following iterative method for the
square root of N:
where x0 is a given initial guess.
(2) Derive a similar iterative method for the third root of N.
(3) Let N=AB, i.e., write N as the product of two numbers A and B. Then a popular formula for an
approximation of is
Show that you can obtain the above formula by iterating the method in (1) twice with x0 = A or B.
What about x0=(A+B)/2?
2. (25 points) Solve the following set of inconsistent linear equations in the least squares sense using the
normal equation method
pf3

Partial preview of the text

Download Final Exam on Numerical Computations | MATH 451 and more Exams Mathematics in PDF only on Docsity!

Next: About this document ...

CSE/MATH 451 Final, Fall 1997

Name: I.D. Number:

1 2 3 4 5 total

  1. (25 points) Let N be a given positive number.

(1) Use Netwon’s method on the equation x^2 - N =0to derive the following iterative method for the square root of N :

where x 0 is a given initial guess.

(2) Derive a similar iterative method for the third root of N. (3) Let N = AB , i.e., write N as the product of two numbers A and B. Then a popular formula for an

approximation of is

Show that you can obtain the above formula by iterating the method in (1) twice with x 0 = A or B. What about x 0 =( A + B )/2?

  1. (25 points) Solve the following set of inconsistent linear equations in the least squares sense using the normal equation method

What is the 2-norm of the minimum residual?

  1. (20 points) Consider the IVP

Apply Euler’s method to the above IVP with constant step size h , write down the iterative scheme, and show that

yn = (1+ h ) n.

What is y (1)? Show that y (1)- yn = O ( h ).

  1. (15 points) Find polynomial of least degree that interpolates the set of data points:

Find another different polynomial which also interpolates the above data.

  1. (15 points) Write a MATLAB code fragment that computes the following matrix

where p is a positive integer, A is square n -by- n , v is a vector n -by-1, and D is diagonal chosen that the columns of A has unit 2-norm. Use the following function header

function K = Krylov(A,v,p)

About this document ...

Next: About this document ... Jinchao Xu