Scribd
Upload
4 views2 pages

HW 9

This document outlines Homework 9 for a Linear Algebra and Probability course, due on November 22, 2024. It includes various problems involving random variables, joint PMFs, covariance, singular value decomposition, and matrix operations. The tasks require calculations and proofs related to statistical concepts and matrix theory.

Uploaded by

Anya Jajodia
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
4 views2 pages

HW 9

This document outlines Homework 9 for a Linear Algebra and Probability course, due on November 22, 2024. It includes various problems involving random variables, joint PMFs, covariance, singular value decomposition, and matrix operations. The tasks require calculations and proofs related to statistical concepts and matrix theory.

Uploaded by

Anya Jajodia
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/ 2

Homework 9

Linear Algebra and Probability, Fall 2024


Due Friday, November 22, at 5 PM

1. Let X and Y be random variables, each taking values 0, 1, 2, and having


joint PMF pX,Y described by the table below.

X
0 1 2
0 1/18 1/9 1/18
Y 1 2/9 1/9 1/9
2 1/18 1/18 2/9

Do the following:
(a) Find pX and pY (the marginal PMFs of X and Y ).
(b) Find the conditional PMF of X given Y = 1.
(c) find the conditional PMF of Y given X = 2.
(d) Find the mean and variance of X and Y .
(e) Find the covariance of X and Y .
(f) Find the corelation of X and Y .
(g) Find E[(X + Y )3 ].
2. Let X be a random variable satisfying
1 1
P (X = −1) = P (X = 1) = , and P (X = 0) = ,
4 2
and define Y from X by setting
(
5 if X = 0
Y =
1 ̸ 0.
if X =

Prove that X and Y are uncorrelated, but not independent.


3. Suppose an urn contains 4 blue and 5 red balls. We will choose 3 at
random without replacement. Let R be the number of red balls selected
and B be the number of blue balls selected. Find the joint PMF of R and
B.

1
4. Let X be an m × n matrix consisting of m features (variables) and n data
points, and let S be its sample covariance matrix. Prove that, if P is an
m × m matrix, then the sample covariance matrix of the transformed data
P T X is P T SP . Hint: Recall that
1
S= BB T ,
n−1
where B is the matrix obtained from X by subtracting off the mean of
each row. Use linearity of expectation to show that the recentered data
matrix for P T X is P T B.
5. Let  
2 0
A = 1 1
0 2
Do the following by hand:
(a) Find the singular value decomposition of A.
(b) Find the rank 1 approximation of A.

6. Let  
4 6 9 12 25 12
3 9 2 5 9 11
A=
 .
2 5 3 4 9 17
8 7 2 3 14 1

(a) Use the numpy package in Python to compute the singular value de-
composition of A. Example code from class is here https://colab.
research.google.com/drive/1uGg_3lgZDRmIG5TO_BYKn4UpMg5dGSLZ?
usp=sharing.
(b) Use numpy to find the rank 1, 2, and 3 approximations of A.
7. Find the trace of the following matrix.
 
3 5 9
2 8 12
4 9 15

8. Suppose the sample covariance matrix of an 2 × n matrix X of data is


 
5 2
S=
2 5

Find an orthogonal matrix P such that the transformed data set P T X is


uncorrelated. What percentage of the total variance is captured by the
first principal component?

You might also like