KEMBAR78
Test Multivariate 1 | PDF | Principal Component Analysis | Linear Regression
0% found this document useful (0 votes)
57 views9 pages

Test Multivariate 1

Uploaded by

a189463
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
0% found this document useful (0 votes)
57 views9 pages

Test Multivariate 1

Uploaded by

a189463
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/ 9

SULIT

UNIVEKSITI KEBANGSAAN MALAYSIA


The National University of Malaysia

PEPERIKSAAN AKHIR
FINAL EXAMINATION
SEMESTER I SESI AKADEMIK 2023 - 2024
SEMESTER JACADEMIC SESSION 2023 - 2024
IJAZAH SARJANA MTJDA DENGAN KEPUJIAN
BACHELORS DEGREE WITH HONOURS
JANUARJJFEBRUARI 2024 MASA :2 JAM 30 MINIT
JANUARY/FEBRUARY2024 TJJvJE:2 HOURS 30 MINUTES
KO1) KURSUS
COURSE CODE A
I
STQS4I 13

S
STQS4J]3
TAJUK KURSUS
AY
MULTIVARIAT CUNAAN
COURSE TITLE
AL
APPLIED MULTIVARIATE
ARAH A N
INSTRUCTION
NM 1. Kertas mi mengandungi EMPAT (4) soalan.

A
This paper consists of FOUR (4) questions.

SA 2. Silajawab semua soalan dalam bukujawapan yang disediakan.

G
Please answer all questions in the provided answer booklet

AN 3. Anda dibenarkan merujuk senarai formula yang telah diperakukan

B
oieh penyeiaras Kursus.

E
You are allowed to refer to the approved formula sheet by the course

TIK coordinator.

SI 4. Jumlab niarkah bagi kertas mi adalali 70.

R
Total marks for this paper is 70.

V
IE
UN

Na Pendaftaran
Matric No.
Set: (dengan perkataan)/in words

Kertas soalan mi mengandungi 8 muka surat bercetak, tidak termasuk muka surat mi.
This question paper consists of8 printed page 's,). excluding this page,

L
I P Lj I V
CWEDTO tSOOO1:flI5 CSL NO. GUS OlIQO
A%0F]

SULIT
SULIT
STQS4 113

1. A researcher wishes to evaluate the impact of diet type (vegetarian, keto,


Mediterranean) and exercise regimen (cardio, strength training, mixed) on various
health outcomes. The health outcomes are assessed based on blood pressure, cholesterol
level, and body mass index (BMI). Data on health outcomes of 20 participants for each
combination of diet-exercise are then recorded.

a. State the appropriate test that suits the need of the researcher. Justify your
answer. (3 marks)

b. Write the hypotheses for the test in (a). (3 marks)

SA
I
AY
c. Explain the essence of this test using the mathematical formulation involved.

AL (2 marks)

NM
A
d. Identify the maximum value of the test statistics in (b) that will deduce that no

A
GS
significant impact of diet and/or exercise on the health outcome.

N
(7 marks)

BA
KE
T
II
RS
V
IE
UN

I SULIT
SULIT
STQS4 113

2. A multivariate linear regression (MLR) model equation is given as Y = XB + K, in


which Y is the dependent variables, X is the predictor variables, B is the regression
coefficients and E is the error.

a. For a model with n observations, p dependent variables and k predictor


variables, show that
i. the expected value of residuals is a zero matrix with dimension n byp.
(5 marks)

ii. A is an unbiased estimator of B . ( 4 marks)

SA
I
b. To employ the MLR model, should the dependent variables or the predictor

AY
variables correlated? Justify your answer. (3 marks)

AL
M
c. Multicollinearity is a problem in a regression model when Vis correlated.

N
AA i. State what is V. (2 marks)

GS ii. How do we solve the problem of multicollinearity? (1 mark)

AN
EB
TIK
SI
ER
NV
I
U

2 SULIT
SULIT
STQS4 113

3. Your friend seeks for help in solving some statistical issues in her research project. She
wants to reduce the number of variables in the data set in order to obtain a new set of
uncorrelated variables, without losing too much information that the original data set
provides.

a. Based on her readings, she is confused between Principal Component Analysis


(PCA), Factor Analysis (FA) and Correspondence Analysis (CA) since all three
are commonly used for the purpose of dimension reduction. Hence, explain the
differences between these three analyses to help your friend's decision in
choosing the most suitable analysis for her project. (9 marks)

SA
I
AY
b. Your friend also struggles to identify whether correlation or covariance matrices

AL
should be used in the dimension reduction analysis. Guide her on how to solve

NM
this issue. (4 marks)

AA
S
c. She then shared her statistical results on employing the PCA to her dataset. The

G
AN dataset consists of measurements on 60 cell nucleus.

EB The variables: radius, texture, perimeter, area, smoothness, compactness

TIK The R output for eigen value & vectors:

SI elgen(I decc2po5ILioa

R
svalues

E
[1] 3.5871231364 1.3027960143 0.8433637695 0.2506001758 0.0155994021 0.0003094198

V
I Svectcrs

UN
1I
1 1121 1,31 1,41 1151 1161
(1,1 -0.5037434 0.1996607 -0.19551641 -0,09393448 0.452443298 0.6742399435
12,1 -0.2231153 0.2681600 0.92541595 -0.14802292 0.002059*64 0.0003250196
(3,1 -0.5114042 0.1$99154 -0.17532024 -0.0319430 0.373989845 -0.7351082385
14,1 -0.5015124 0.1968141 -0.19720094 0.11159801 -0.607123540 0.0335019586
15,1 -0.2026141 -0.7609319 0.07378165 -0.61154118 0.021825400 -0.0044898821
16.1 -0.3773072 -0.4948732 0,11491905 3.75956358 -0.050286672 0.0546922298

After discussing with her supervisor, she decides to proceed with a 2-factor
model. Help her to compute the communalities and the specific variance of each
variable using Principal Component Method. (8 marks)

3 SULLT
SULIT
STQS41 13

d. Since the communalities and specific variances can also be estimated using
Maximum Likelihood and Principal Factor methods, suggest what should be
done by your friend to identify the best estimation method. (2 marks)

e. Perform the calculation that you suggest in (d) for estimation obtained by
Principal Component Method. (2 marks)

4. Multivariate analysis not only deal with the issue of dimension reduction but also
involves data categorization. To reduce the dimension of data in order to be visually
visible, multidimensional scaling is one of the popular choices. For categorization

SA
I
purposes, cluster and discriminant analyses are often employed.

AY
a. What is the purpose of STRESS value in non-metric dimensional scaling?

AL
Explain how decision can be made from this value. (3 marks)

NM
A
b. Explain the differences between single, complete and average linkages in

A
S
hierarchical clustering algorithm.

G
(5 marks)

AN
B
c. Azran wants to classify graduated university students into working and
E
TIK unemployed categories based on their current job status. He collects sufficient

SI data to estimate the density functionfj(x) andf2 (x) for population ,r; (working)

E R and 7r2 (unemployed). Given that 20% of the graduated students are from

V
I population 7t2, and the density function value of a new obs&vation, xo is

UN f;( x & =0.3 andf2(x&04,


i. classify the new observation into population 7ri or ff2. (3 marks)

ii. if the prior probabilities of these two populations are equal, classify the
new observation into population 7ri or ff2 . (4 marks)

"GOOD LUCK"

4 SULIT
S U LIT
STQS41 13

STATISTICAL TABLE

0.5Cm 038C0 0.508* 0.5120 O$t6O 0.5109 03239 03279 0.8319 05359
0,5358 03438 0.5478 0.5517 0.5557 0.5596 0.5636 0.5675 0.5714 03754
0,5793 015832 0,5871 0,5910 0.5948 03987 0,6026 0.6064 0.6103 0.6141

A
I
0.6179 0,6217 0.6255 0.6292 0,6311 06368 0,6406 0.6443 0,6460 04517

YS
A
oinsá ann 6M74 o,nsi o.nn 0.72. 01454 0,748 0,7518 0.7549

AL
0.1580 0.7612 0.1647 0.7673 0.7704 0.7731 0,7764 0.7194 0J7813 03852
047581 0.7910 0.1939 0.7967 0,1995 048023 0.8051 o;eola 02106 02163

0.8413

NM0.8438 0.8481 0,8485 0.8508 0.8531 0.8554 0.8577 0.6599 0.8621

A
0,8643 0.8885 0.86*6 0.8708 0.8729 0.6749 0.8770 0.8790 0.8810 0,8830

A
0.49 0.8869 0.8888 0.8907 0.8925 0.8944 0.8962 02980 0.8997 0,9015

S
0.9232 0.9049 0.9066 0.9087 0,9099 03115 03131 03147 0,9162 0.9177

äs32

NG 0.9345 0.9357 0.4370 0.9382 0.9394 0$4ffi 0,9418- '0313U 113441

A
0.9-432 0.9463 0,9474 0.9465' 0.9495 03305 0,9516 0.9323 tSSSS -0.9545

B
0S554 0.9564 03573 03582 05591 0.9599 05608 Q.%16 0.9623 .0.9633

E
0.9641 03649 -0.9656 0.9664 &4s71 03573 0.9686 049692 &97a 09708

K
{7,97j9 0.9726 03732 05738 0.9744 0S750 0,97% 0!! 0 0,9167

I
0.9713 .
0.93 03808 0.9812 0,9817

T
0,9777 03778 0.9783 03783 0.9793 0.9798

I
0,9326 0.9830 0,9834 0.9838 0.9842 03848 0.9960 03854 0,9857

S
03821
0.9875 0.9878 0.9881 038*6 0.9837 03890

R
0.9861 03865 03888 0.9621
03911 0.9913 0.9916

E
0.9893 03696 0.9898 0.9901 0,9904 0.9906 03909

V
I om3* 0s94o 0.9941 0th45 0,9945 0,9948 0,9948 03949 0.9951 0,9952

UN
0,9953 0.9955 0,995 0.9952 03959 03950 03961 0.9962 0.9953 03964'
63965 aige 03967 03966 0.9989 03970 0.9971 03972 0397* 03974
O 9974 *3975 0.9916 049977 03977 0,9976 03979 03930 0.9980 OSSM

togs 0.9987 0.9988 03988 - 03929 0.9980 03989 03990 03990


0,9990 0.9991 0.9991 0,9991 0.9992 0,9992 0.9952 03992 0.9903 0.9993
03993 03993 0.9994 0.9994 0.9994 0,9991 09994 0.9995 0.9995 0.9995
03995 03995 0.9995 0.9996 0.9998 0,9998 0,9996 0.9996 0.9996 0.9997
0.9907 0.9997 0.9997 0.9997 0.9987 0.9997 0.9997 0.9997 0.9958 0.9998

5 SULIT
SULIT
STQS41 13

curn.prob to (ji tso is, fit t.,, tin tsa tin t, tj,s
øne.taII 0.50 0.25 0.20 0.15 0.10 0.05 0.07.t 0.01 0.005 0.001 0.00051
4I'M'.
nfl ncn nan
-. Am
-.-.- 0.20
--.--. 0- .10- 0,05 0.02 0.01 0.002 0.001
Cf
1.376 1.963 3.078 6.314 12.71 31.82 63.66 318.31 636.62
I 0.000 1.000
0.816 1.061 1.388 1.686 2.920 4.303 6.955 9.925 22.327 31.599
2 0.000
0.978 1.250 1.638 2.353 3.162 4.541 5.841 10.215 12.924
3 0.000 0.765
0.941 1.190 1.533 2.132 2.778 3.747 4.604 7.173 8.610
4 0.000 0.741
0S20 1.156 L478 2.9 L57135 t032 5M93 8.869
57
1,440 -1
0.711 0.896 119 1.415. 1.895 2.386
1 2.998 3.499 4.785 5,4081
7 0.000
6 0000 0.706 0.889 1.108 1.397 1.880 2.306 2.896 3.355 4.501 5.041
9 0.000 0703 0683 1.100 1.383 1S332.282 2:821 3250 4297 4.781-
10 O.000. -372 1.812 2228 &7_64 3.169 4.11,111,44
II 0000 0.697 0.976 1.0881,363 1.798 2.201 2.718 3.108 4.437
4.025
12 0.000 0.695 0.873 1.0831.358 1.782 2.179 2.681 3.055 4.318
3.930
1.0791.350 1.771 2.160 2.650 3.012 4.221
3.852

A
13 0.000 0.694 0.870

SI
14 0.000 0.692 0868 1.0761.345 1.781 2,145 2.624 2.977 4.140
3.787
15 0.000 0,691 O866 1.0741341 1.753 2 Ir
30l
-••-*b4%
2.947

Y
746

A
0.689 0.863 1.069 1.333 1.740 2.110 2.567 2.898- 3.648 3.965

L
17 0.000
18 - 0.000 0.688 -0.862 1067 1330 1.734 2.101 2.552 2.878 3.610

A
19 . 10.000 0.688 0.861 1.068 1.328 1.729 2.093 2.539 2.861 3.579 '3883

M
L_20 Q00j8sL...0.%L_l ,04 . ..1 325_1 .26 _2.0_?_LL3AQ2_3.65Q
0.000 0686 0.859 1063 1323 1.721 2080 2.518 2.831 3.527 3.819

N
21
22 0.000 0.688 0.858 1.061 1.321 1.717 2.074 2.508 2.819 3.505 3.792

A 23 0.000 0.685 0.858 1.060 1.319 1.714 2.069 2.500 2.807 3.485 3.768

A 24 0000 0.685 0.857 1059 1.318 1.711 2.064 2.492 2.797 3.467 3.745

GS 25 000 0.684 0856


rt 4% 4;
N
27 a.000 - 0.684 0. 855 1 057 1.314 1.703. 2.052 2.473 2.771 3.421 . aS9OJ

A
28 0.000 683 0.55
8 1.056 1.313 1101 2.048, 2.467 2.763 3.408 3.6741

B
29 - 0.000 - 03 68 0.854 1.055 1.311 1.69 2.045: 2.462 2.756- 3.396 :?.659

E
L ............30 _o.poç ._.0fi8p_0 541.05_j.310 _107 2.042 Z451

K
40 0.000 0681 0851 1.050 1,303 1.684 2.021 2.423 2.704 3.307 3.551

I 60 0.000 0.679 0.848 1.045 1.296 1.671 2.000 2.390 2.660 3.232 3.460

T
I
80 0.000 0.678 0.846 1.043 1.292 1.664 1.990 2.374 2.639 3.195 3.416

S
100 0.000 0.877 0.845 1.042 1.290 1.660 1.984 2.364 2.626 3.174 3.390

R
1000 0.000 0.675 0.842 1.037 1.282 1.646 1.982 2.330 2.581 3.008 3.300

E 0000 .. 0874 0842 1036 1282 1645 1.960 2326 2578 3090 32911

V
Z

I 0% 50% 60% 70% 80% 90% 95% 98% 99% 99.8% 99.9%

UN
Confidence Level

SULIT
SIJLIT
STQS4I 13

• -- F-table of Critical Valoes ola=t05 for F(dfl. df2), -


Dfl=1 2 3 4 5 6 7 2 '9 10i2 152024304060120
11F2=1 168 45 89930 28521 22438 23016 23399 23677 23S2 240.54 24888 24.°l 243.95 24801 24905 23010 231.14 25220 25325 25438
'15 SI 1900 1916 19.25 19.3019'33 1935 1937 1938 1940 19.41 t9.43 1945 19.45 89.46 1947 1948 149i10
3 1013 9.55 928 9.12 9.01 894 8.89 8.85 S Si 819 8.74 870 8.66 864 862 8.59 857 8.55 8.53
4 131&94 659 639 616 6.16 -609 404 600 596 -591 536 5.80 577 S75 522 5.69 5.66563
5 661 5.79 41 519 50) 4.95 488 4.82 477 474 468 462 4.56 453 4.50 446 443 4.40 431
6 5- 514 4769 4.39-j.28 421 4.15 4.10406 4.00.394 3.87 384 3.81 3.77 374 370 3.67
7 5.59 4.74 435 4.12 397 3.87 3.79 3.73 368 3.64 357 321 344 341 3.38 334 330 327 323
-. .•..-.
S. 4.46 4:07 3.84 br 338 330 144 339 335 128322.3.15 J- 12 , 3 -GS3.04 301 2.97 2.93
9 5. 82 4.263.863633.483373293233.183143.073012942.902 86 2.83 2.79 2.7S 2.71
496 414 s.iiT4s 3.33 322 334 107 iM298 231 285 27? 2.74 270 2.6C242 2.58 2.54
11 484 3.98 3.9 3.36 3.20 109 301 2.95 290 285 2.79 272 2.65 261 257 253 249 245 2.40
iz 475 Th89 349324 3 11i00 Ir7iS '280 275 269262 _2-47_2A3,,
2'4 251 2.38 23430
13 4.67 381 341 318 303 2.92 283 2.77 271 267 260 2.53 2.46 242 238 234 230 2.25 221
14 4.60374 334311 l96-285 1762.70 265 2.40 233 14'5 -2,9 235 131 3i72.22 218 2.13
15 4.54 368 3.29 306 2.90 279 2. 7 1 264 2.59 2.54 248 2.40 233 229 225 120 2.16 211 2.07

A
234 2.49 2.42 2.35, 2.28224249215 231"M1.06, Z--Ol

I
>16 4.49 , 33324 3.01 285 1S9ZIll,

S
17 4.45 359 3.202% 2.81 210268 2.55 2.49 2.45 238 2.31 223 219 2.1 210 206 201 836

Y
441 335 , 16 2.93 U7:2oc38 2.51 24624i 2.34 227 2.19 215 2.11 206 202 L97 1.92
19

LA
4.38 32 313 290 2.74263 254 2.48 2 42 2.38 231 2.23 216 211 2.07 203198 133 188
435 349 3.10 287 2.71 260 2.5 2.45 '239 23.5 228 2.20 212 108 2.04 149 135 1 90184

A
20
21 4.32 3.47 3.07 284 2.63 257 249 242 237 232 2.21 218 210 203 201 1.96 192 187 1.81

23

NM
--22 4.30 3.44-3.05 2.82 2.66 235 >2.46 2.40 234 230 '2.23 2.15207 2.03 138 t94I.89 1:84 1.78
428 342 3.03 280 2.64 253 2.44 237 2.32 227 120 213 20' 201 196 198 1.86 181 176

A
24 4.26 3.40 101 uS 2.62 231i42 uS 2302.25 2.18 211 '.2.03 1.98 1.94 1.89184 1.19 173

A
25 424 339 299 276 2.60 149 2.40 234 228 224 2.16 209 2.01 196 192 187 182 1.77 1.11

S
26 413 '3.372S&:274 2392.4T 2.39 1321.27 2.22 2.15 207 1.99 195 1.90 L85 1.80 175 169

G
27 411 335 296 2.73 2)7 2.46 237 231 225 220 213 206 1.97 193 182 184 179 1.73 167

N
23 420 334 - 1235 2.71 2.56 145 2.36 229 224 219 132 2.04 136 391
117- 1.82 l.77 L7IT>1.65

A
1

29 418 3.33 293 2.70 2.55 2.43 2.35 228 2.22 218 2.10 20, 1.94 190 I 85 1.61 115 170 1.64

EB O 4.11 3.32 2.92 2.MJ 253 242 2iit 2.27r2.ui 216 209 -201 .193 189 1.84 1.79 i74 L68 1.62
408 3.23 2 84 2,451 245 2.34 225 2.18 212 208 206 192 1.84 179 174 169 164 1.53 1.51

K
40

I
60 400 315 2.76 253 237 > 22 217 210 204 199 192 184 175 170 165 139 153 147139

T
392 3.07 268 2.45 229 218 209 2.02 196191 1.83 175 166 161 1.55 150 143 1.35 1.25

I
120

S
386,300 2.60 237 22.1 2102.01 194 1,88 1.83 1..7s 1.67 837 132 1.46 139-132 1.22 tOO

ER
NV
I
U

SULIT
SULIT
STQS4 113

Right-tall

critical

Chi-Square Right-Tail Probability a xi


flF o qqc 0.99 0.915 0.95.., i 0.9 0.1 0.05 ' 4 0.
0.025 0.005
--i- - - y. - 00010004I0M161 27063..841i50246..035 7179
-u- ---- . -' 0 , 103 0211 4605 5991 7378 '9210 10597
12 0010 0020 0051
..................... ... . ...... .
3 0.072 . 0.115 0.216 0352 0.584 6.251 7.815 9 .348 11345 11838
4 0207

SA
I
0291.,.-04S4 .-...... 0711 1064 7 , 779 9488 11143 277
13-,--
-. ....-'---.- 14860

Y
5 0.412 0354 0.831 1.145 1.610 9.2.36 11.070 112.833 15.086 16.150
L"6 t

LA
, 0616_ 0872 1237 L) 1204 9645_12592..J4449 1t8i218548
14.067 : 16 .013 18.475
A
7 0589 1.239 1.690 E 2.167 2.833 12.017 20.278
.........

M
1344 1646 2.180 2 . 733 3490 .13 . 3,62 15507 1753 2009021955

N
9 1.735 2.088 2.700 3.325 4.168 1 14.684 16.919 19.023 21.666 23389

A
110 1156 2558 3247 394w 4865 15987 18307 20483 23209 25188
11
SA 2.603 . 3.053
- ...
3.816 4.575
. 5.578 17275 19.675 21.920 24.725 26.757

G
12;.! 3074 .. 3571, ^8_549 2^1026 23-337 26 ,
2 17 28 300 4

N 4.107 5.009 . 5.892 7.042 19.812 11.362 24.736 27.688 29.819

A
13 3.565
- ......

B
i4 ., 4075 4 __ 0

E 6.262 7261 8.547 22.307 24.996 27.488 : 30.578 32.801

K
15 4.601 1 5.2.9

I
S16 5142 Tssiz 6908- 7962. 9312 23542 26296 28845 12000 : ... 34267

ST
I 17 5.697 6.408 7.564 8.672 10.085 24.769 27.587 . 30.191 33.409 35.718

R
18 t265 lots -R231 /9390 10865 25989 28869 31526 34805 31156
:

E 8.907 10.117i 11.651 27.204 30.144 . 32.852 36.191 38.582

V
19 6.844 7.633

I
3999j1
7743477 -1 93911 10851 12443 28412 31410 34170 .'37i66

UN 21
(21
8.034
8643
8.897 1 10.283 11.591 13.240 . 29.615 32.671 35.479 . 38.932 .....1 41.401
-: 12338 14041 30813
9542-- 1082.-r------r--------r-- 33924 36781 . -.40289
---,---......................... ......... 42796 1
44.181
23 9.260 10.196 : 11.689 13.091 . 14.848 i 32.007 35.172 38.076 41.638
9886 ........................
10-856 ... 12401 13 848 15 659a,.. 33 196 36415 39364 42980 .. 45559
L24 ..-_-j-._.,.f-.-.---_
25 10.520 11.524 13.120 14.611 16.473 34.382 37.652 40.646 44.314 46.928
C2 L- 11160 12 198 13 j!0. 17,292 1 , 35-51: 2f l 5, i1923 482% i
27 11.808 12.879 14.573 16.151 18.114 36:741 40.113 43.195 . 46.963 49.645
12 461 ,13 565 ,,, J5308 ,, j6 928_ 39,,,,j,2L& j1j37 44461 4817S 50 J
29 13121 14256 16047 17-7081 19168 39087 42557 45722 49588 52336
..
13787 14953 s 16 791Ti'TZ3'205? 40256 43773 ..,46979 50292.-.53672
40 20707 22164 24433 26509 29051 51805 55758 59342 63691 66766

60
E!E
80
35.534
L'^
37.485 40.482
51Y1J
43.188 46.459 74.397 79.082

51.172 53.540 1 57.153 1 60.391 j 64.278 96.578 101.879: 106.629: 112.329 116.321
P2°J
83.293 83.379 91.952

90 < 59.%196 )1 7S465 67 9126 7fl91 _4O65Ql3 145Jl18- 1364 1112829


100 67.328 70.065 74.222 . 71.929 82.358 i 118.498 t 124.342 129.5611 135.807 : 140.169

SULIT

You might also like