0% found this document useful (0 votes)
36 views5 pages

Math 425 Multivariate Data Analysis - Kabarak University

Uploaded by

billtravis1998
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
36 views5 pages

Math 425 Multivariate Data Analysis - Kabarak University

Uploaded by

billtravis1998
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

KABARAK UNIVERSITY

UNIVERSITY EXAMINATIONS
MAIN CAMPUS
SECOND SEMESTER 2022, ACADEMIC YEAR
EXAMINATION FOR THE DEGREE OF BACHELOR OF SCIENCE:
ECONOMICS AND STATISTICS, AND ECONOMICS AND
MATHEMATICS

MATH 425: MULTIVARIATE DATA ANALYSIS

STREAM: Y4/S2 - REGULAR TIME: 2:00-4:00PM


EXAMINATION SESSION: MAY - AUGUST DATE: 29/07/2022

INSTRUCTIONS TO CANDIDATES
1. Answer Question 1 and any other two questions in the answer booklet provided.
2. Do not write on your question papers. All rough work should be done in your answer
booklet.
3. Clearly indicate which question you are answering.
4. Write neatly and legibly.
5. Edit your work for language and grammar errors.
6. Follow all the instructions in the answer booklet

As members of Kabarak University family, we purpose at all times and in all places, to set apart in one’s heart,
Jesus as Lord. (1 Peter 3:15)
Kabarak University is ISO 9001:2015 Certified
Page 1 of 5
QUESTION ONE [30 MARKS]
(a) Explain the following terms
(i) Principle component analysis (1 mark)
(ii) Random vector (1 mark)
(iii) Multivariate data (1 mark)
(b) Let x be a random vector having the covariance matrix

[ ]
4 1 2
Σ= 1 9 −3
2 −3 25
Obtain

( )
1
(i) Square root of Σ=¿ V 2 (2 marks)

( )
1 −1
(ii) Inverse of the square root Σ ¿ V 2 (1 mark)

(iii) Correlation matrix ρ defined by

( ) Σ (V )
1 −1 1 −1
ρ= V 2 2 (2 marks)

(c) Let x=[ 1 ,3 , 2 ] and y= [−2 , 1 ,−1 ] find


(i) The length of x (1 mark)
(ii) The angle between x and y (3 marks)
(iii) The length of the projection of x and y (1 mark)

(d) Let A= [ 3
−√ 2
− √2
2 ]
(i) Is A symmetric? Give reason (1 mark)
(ii) Show that A is positive definite (4 marks)

As members of Kabarak University family, we purpose at all times and in all places, to set apart in one’s heart,
Jesus as Lord. (1 Peter 3:15)
Kabarak University is ISO 9001:2015 Certified
Page 2 of 5
(e) Consider the following n=7 observations on p=2 variables

x1 3 4 2 6 8 2 5

x2 5 5.5 4 7 10 5 7.5

(i) Compute the sample means x 1 and x 2 and the sample variances S11 and S22 (4 marks)
(ii) Compute the sample covariance S12 and the sample correlation coefficient r 12 and
interpret these quantities (5 marks)
(iii) Display the sample mean array x , the sample correlation array R and the sample
variance-covariance S12 (3 marks)

QUESTION TWO (20 MARKS)


(a) Let x be a p-variate random vector with mean vector μ and variance covariance
matrix Σ , show that E ( X X ' )=Σ+ μ μ ' , hence show that
E ( X ' A X ) =trace ( A Σ )+ μ ' A μ where A is a symmetric matrix of constants.
(8 marks)
(b) Find the symmetric matrix A for a quadratic form
2 2 2
Q ( X 1 , X 2 , X 3 )=9 X 1+16 X 1 X 2+ X 2+ 8 X 1 X 3 +6 X 2 X 3 +3 X 3. Hence obtain the

() ( )
6 1 2 5
expected value of Q ( X 1 , X 2 , X 3 ) and E( X ' AX) given that μ= 7 and ∑= 2 4 3
3 5 3 9
(12 marks)
QUESTION THREE (20 MARKS)
a) In an experiment involving two correlated variables, the following sample statistics were

obtained: X = [ 10.00
10.00 ] S=
[ 0.7986
0.6793
0.6793
0.7343]. Determine

(i) Principal components (8 marks)


(ii) Variance of each principal component (3 marks)
(iii) Percentage of variance explained by each principal component (3 marks)
As members of Kabarak University family, we purpose at all times and in all places, to set apart in one’s heart,
Jesus as Lord. (1 Peter 3:15)
Kabarak University is ISO 9001:2015 Certified
Page 3 of 5
b) Assume x ' =(x 1 , x 2 , x3 ) is normally distributed with mean vector μ=(1 ,−1 , 2) and

[ ]
4 0 −1
variance matrix Σ= 0 5 0 . Find the distribution of 3 x 1−2 x 2 + x 3 (9 marks)
−1 0 2
QUESTION FOUR (20 MARKS)
(a) Find the maximum likelihood estimators of the mean vector μ and covariance matrix Σ based
on the data matrix (6
marks)

[ ]
4 1
x= −1 3
3 5

[ ]
1 9 10
4 12 16
(b) Given the data matrix x= 2 10 12
5 8 13
3 11 14

Define X c =X−1 x ' as the mean corrected data matrix.


(i) Obtain the mean corrected data matrix (4 marks)
(ii) Obtain the sample covariance matrix (4 marks)
(iii) The generalized variance and hence verify that columns of mean corrected data
matrix are linearly dependent. (3 marks)
(iv) Specify a vector a ' =[a1 a2 a3 ] that establishes the linear dependence
(3 marks)
QUESTION FIVE (20 MARKS)
(a) Let x be a random vector having the covariance matrix

[ ]
25 −2 4
Σ= −2 4 1
4 1 9
1
(i) Obtain the population correlation matrix ( ρ ) and V 2 (6 marks)

As members of Kabarak University family, we purpose at all times and in all places, to set apart in one’s heart,
Jesus as Lord. (1 Peter 3:15)
Kabarak University is ISO 9001:2015 Certified
Page 4 of 5
1 1
(ii) Multiply your matrices to check the relationV 2 ρV 2 (4 marks)

(b) Let A= [−29 −26 ]


(iii) Is A symmetric? Give reason. (1 mark)
(iv) Obtain Eigen value (3 marks)
(v) Show that A is positive definite (6 marks)

As members of Kabarak University family, we purpose at all times and in all places, to set apart in one’s heart,
Jesus as Lord. (1 Peter 3:15)
Kabarak University is ISO 9001:2015 Certified
Page 5 of 5

You might also like