Scribd
Upload
138 views1 page

fmnf05 2020 Ex3

This homework assignment involves analyzing matrices and their properties. It contains 4 questions: 1) Analyzing an n×n matrix including computing its infinity norm, solving systems of equations, and plotting condition numbers and relative errors. 2) Plotting unit vectors in 1-norm, 2-norm, and infinity norm spaces. 3) Plotting the effect of multiplying unit vectors by different 2x2 matrices, identifying largest vectors, eigenvectors, and explaining the results. 4) Computing eigenvalues, eigenvectors, and condition numbers by hand for 2x2 matrices, and using an power method algorithm to approximate eigenvalues within tolerance for the matrices.

Uploaded by

Bengt Hörberg
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
138 views1 page

fmnf05 2020 Ex3

This homework assignment involves analyzing matrices and their properties. It contains 4 questions: 1) Analyzing an n×n matrix including computing its infinity norm, solving systems of equations, and plotting condition numbers and relative errors. 2) Plotting unit vectors in 1-norm, 2-norm, and infinity norm spaces. 3) Plotting the effect of multiplying unit vectors by different 2x2 matrices, identifying largest vectors, eigenvectors, and explaining the results. 4) Computing eigenvalues, eigenvectors, and condition numbers by hand for 2x2 matrices, and using an power method algorithm to approximate eigenvalues within tolerance for the matrices.

Uploaded by

Bengt Hörberg
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/ 1

FMNF05 Spring 2020 - LTH - Alexandros Sopasakis

Homework Assignment 3.
Latest due date: midnight - February Sunday 9th
Possible bonus of 0.1 if 75% correct (≥ 18 pts.)
Please include all requested figures, programs and explanations in the PDF report!
1. (6 pts.) Consider the n × n matrix A with entries Aij = 5/(i + 2j − 1).
• (2 pt.) Compute the infinity-norm of the matrix by hand (show all steps in your
report) and then verify it by writing an algorithm with n = 6, 7, . . . , 14, 15.
• (1 pt.) Set x = [1, . . . , 1]T and let b = Ax. Now solve the system Ax = b for x
(as if you did not know it) for n = 6, 7, . . . , 14, 15 using the backslash operator
(Matlab) or solve(A,b) (Python).
• (1 pt.) Create a plot of the condition number (y-axis) versus n (x-axis) for each
n above. You may use Matlab/Python to compute the condition number.
• (2 pt.) Create another plot of the relative error (y-axis) versus n (x-axis) until you
identify the value of n after which the solution have no correct significant digits.
What is that n value? Does that value depend on your particular computer?
No need to include your programs for this question.
2. (4 pts.) Plot the set of all unit vectors v in R2 in 1-norm, 2-norm and the infinity norm. (each
worth 1 pt). Include your code in your report (1pt).
Please only plot a point representing where a vector ends instead of the full vector.
3. (10 pts.) Consider the following 2x2 matrices:
     
1 2 1 4 Create a singular 2x2
A1 = A2 = A3 =
2 1 −4 1 matrix of your choice
• (3 pts.) For each set of points in Question 2 above plot the corresponding A1 v, A2 v
and A3 v. In total 9 different plots should be created. Make sure to also include
the original set of points v, from Question 2, in each plot;
• (3 pts.) mark the vector with largest 1-norm, 2-norm and infinite norm;
• (1 pts.) mark also the eigenvectors (if possible);
• (2 pts.) explain, based on the theory, why you marked the vector above as the
eigenvector-full explanations required for credit;
• (1 pts.) include your code for all questions above with your report.
4. (4 pts.) Consider matrices A1 and A2 as given above.
• (1 pt.) Compute by hand the eigenvalues as well as eigenvectors for each of those.
• (1 pt.) Also compute by hand the corresponding condition number of each matrix.
• (2 pts.) Program the power method and approximate both the eigenvalues within
a maximum tolerance of 10−4 (per eigenvalue) for each matrix A1 and A2 .
Report the number of iterations to approximate the eigenvalues for each matrix
A1 and A2 . Include your algorithm in your report.

You might also like