Scribd
Upload
2K views3 pages

Multivariate Analysis Practice Question and Solutions

This document contains solutions to tutorial problems involving multivariate analysis. It covers proofs involving eigenvalues and eigenvectors, finding the eigenvalue decomposition of matrices, and calculating determinants, traces, powers and roots of matrices using their eigenvalue decompositions. Matrix properties like orthogonality and positive definiteness are also discussed.

Uploaded by

Janice Txinyi
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
2K views3 pages

Multivariate Analysis Practice Question and Solutions

This document contains solutions to tutorial problems involving multivariate analysis. It covers proofs involving eigenvalues and eigenvectors, finding the eigenvalue decomposition of matrices, and calculating determinants, traces, powers and roots of matrices using their eigenvalue decompositions. Matrix properties like orthogonality and positive definiteness are also discussed.

Uploaded by

Janice Txinyi
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/ 3

MH4501 Multivariate Analysis

Tutorial 1 Solutions

1. Proof:

(a) QQT = Ip . Then 1 = |I| = |QQT | = |Q||QT | = |Q|2 = |Q| = 1.


(b)
Aui = i ui = uTj Aui = uTj i ui

Auj = j uj = uTi Auj = uTi j uj

Since matrix A is symmetric, i 6= j , then

i uTj ui = j uTi uj = uTi uj = 0.

(c) We have the eigenvalue decomposition A = QQT ,

det(A) = |A| = |QQT | = |Q| || |QT | = |Q|2 || = || = 1 2 p .

(d)
p
X
T T
tr(A) = tr(QQ ) = tr(Q Q) = tr() = i .
i=1

4 2
2. A =
2 3

(a)

|A I| = (4 )(3 ) 2 = 0
= 1 = 5, 2 = 2.

Let x = (x1 , x2 )T .

1

If Ax = 5x, then 4x1 2x2 = 5x1 and 2x1 + 3x2 = 5x2 . It follows that x

satisfies x1 + 2x2 = 0. Let us take x1 = 1 and x2 = 2/2. Then


x 6 1 6
u1 = = x = .
kxk 3 3 3


If Ax = 2x, then 4x1 2x2 = 2x1 and 2x1 + 3x2 = 2x2 . It follows that x

satisfies 2x1 2x2 = 0. Let us take x1 = 1 and x2 = 2. Then


x 3 1 3
u2 = = x = .
kxk 3 3 6

It follows that
1 6 3
U = [u1 , u2 ] =
3 3 6

and since U is an orthogonal matrix



1 6 3
U 1 = U T =
3 3 6

(b) Omitted.
(c) You need to use A5 = U 5 U T . You may show this directly or use the result that
the eigenvectors and eigenvalues of A5 are u1 , u2 and 51 , 52 . So we have
5
1 6 3 5 0 6 3
A5 = U 5 U T =
9 3 6

5
0 2 3 6

2094 1031 2
=
1031 2 1063

(d) You need to use A1/2 = U 1/2 U T . So we have



1 6 3 5 0 6 3
A1/2 = U 1/2 U T =
9 3 6 0 2 3 6

1 2 5 + 2 2 10
=
3 2 10 5+2 2

2
(e) Omitted.

3. (a) |A I| = (2 )(4 ) 16 = ( + 6)( 4) and hence eigenvalues are 1 = 4


and 2 = 6.
A is not positive definite because 2 < 0.
(b)

1 12 5 5 12
DDT =
169 12 5 12 5

1 169 0
= =I
169 0 169

Hence DT = D1 and so D is orthogonal.


(c) We have
|A| = 24, |B| = 14, |D| = 1

and therefore
|AABDDD| = |A||A||B||D||D||D| = 8064.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy