Birla Institute of Technology & Science, Pilani

MATH F432 (Applied Statistical Methods)

First Semester 2021-2022
Comprehensive Examination (Open Book)
Max. Marks: 70
Date: December 11, 2021 (Saturday)
Time: 180 Minutes

and when required. Assunptions

usual meaning. Define random variables as
Note: Notations and symbols have their on fresh page. Moreover,
answer each
the beginning. Start new question
made if any, should be stated clearly at
continuation.
Subpart of a question in

normal variates with their mean and standard deviations oj and d2, respectively.
et X and Y be two independent -3X > 1) 1, then compute j. 10
6)+ P(Y
=

1fP(2X +4Y <9) 1 and P(2X-4Y

10)+ P(3X +Y
=

new synthetic fiber that

will be used to make
needs to investigate the tensile strength of a
A product engineer knows that strength is affected by
is experienced in this type of work and
cloth for an application. The engineer the new fiber. He suspects that increasing
the
used on the blend of materials for
the percent of cotton (by weight) that is, to test whether the
at least initially. To conduct this experiment,
cotton content will increase the strength, mean tensile strengths,
the
with diferent percent weights of cotton have equal
5 blended fbers (A, B. C. D and E) test at 57 significance level. 10
observations. Pertorm a suitable non-parametric
engineer considers 25

Weight percent of cotton

A(15%) B(20%) C(25%) D(30%) E(35%)

14 19
2
17 18 25
15 12 18 22
19 19 15
11 18
18 19 23
9
a statistical
a civil engineer has used
variables related to bridge construction,
T o analyse the interrelationship of some and principal
to process the data according to
its built-in defaults (standardization
program of factor analysis
extracted two factors and produced the following factor loadings.
component estimation). It has

Unrotated factor loadings

Variable
Factor Factor 2
0.70 0.47
TIME (Design time in days)
DAREA (Deck area of bridge in sq.t.) 0.74 0.44
CCOST (Construction cost in dollar) 0.83 0.25
DWGS (Number of structural drawings) 0.50 0.65
LENGTH (Length of bridge in feet) 0.94 0.16
0.80 0.25
SPANS (Number of spans)
DDIFF (Degree of difficulty of bridge design) 0.15 0.93

percentage of its variance that is explaned by the tactors.

aalculate the communality of each variable and the

) Compute the eigen values of factor 1 and factor 2.

I f the sum total of eigen values is 6.0, compute the percentage of the total varianee that is explained by each
factor and by both factors jointly.
T Now the engineer performs varimax rotation as a routine job of factor rotation. hat 1s the purpose of
factor rotation anyway? (i) What would be the communality of each variable now?( i ) What percentage of
the total variance will now be explained by each factor and by both factors jointly? 10
The low birth weight (LBW) in babies is a serious problem even in developed countries. In this regard, data was
obtained fron 189 women who gave birth during a particular period at the Bayside Medical Center in Springfield
Masaechusettn, he predictors are weight ofthe mother at her last menstrual period (w), age of the mother (a), and
whether the mother had regularly
sinoked luring pregnaney (s). (i) With proper justihcation, formulate a logistic
regression model lor tlhe above problem. (PSuggest, without any computation, how to test whether the variable
related t0 moking is inlormative or ot in analysing L3W
5
ASAaple of 200 ROM computer chips was selected on each of 30 consecutive dlays, and the number of nonconforming
chips on each day was as follows: 10, 18, 24, 17,37, 19, 7, 25, 11, 24. 29, 15, 16, 21, 18, 17, 15, 22, 12, 20, 17, 18,
12.2430, 16, 11, 20, 14, 28. Construct a p chart and examine it for any out-of control points. 5
While carrying out the signilicance of a discriminant function model that diseriminates the effect of Covid-19 on
smokers and non-smokes, a one-way ANOVA was conducted. Part of the summary table is provided below. If the
eigenvalue for the above estimated discriminant function is 1.5, compute Wilk's lambda and the missing entries of
the table. 5
Sum of Squares d.f. Mean Square P-valúe
Between Groups
Within Groups
Total 100 25

. F o r a randomsample of size n from a population having its mean and standard deviation a, prove or disprove
that T= X X is an unbiasecd estimator of a . Show all computations.
5
Using Manhattan distance metric, perform agglomerative clustering to segnent the below cities into two diferent
groups.
5
Cities Latitude (°N) Longitude (°E)
5 20
10 17
14 28
11 15
20 15

9. A random sample of size 4 is taken from a normal distribution with unknown mean
u and variance
a > 0. To test
Ho:u=0 against H1: p<0, the following criterion is used: "Reject Ho if and only if X1 + X2 + X3 + X4 <20.
Find the value of a so that
the significance level of this test will be closed to 0.2. Also. find the power of the test
when f =-0.2.
[5
10. Fill in the blanks and short-answer type questions
(a) Generally, if the distribution of data is skewed to the
left, the ... is less than the median, which is
********

often less than the

(b) lf the umber of number
of estimated parameters in the model
function for the model is
is, 3, and the maximum value of the likelihood
156.2, then the AIC value of the model is .
(c) .

tests the hypothesis that your

[1
correlation
. . .

matrix is
identity matrix.an
1n
r .

statstics 1s a transformation applied to a non-stationary time-series in order to make it

n the mean sense
(viz., to remove the non-constant trend), but having stationary
of the variance/autocovariance. nothing to do with the non-stationarity
s o s s t 1 8 Used to test the stationarity [1
a unit root 1s present in a time
(trend-stationarity) of the time series with the null hypothesis that
series sample,
When is a
multiplicative time-series model recommended
What is the major difference
between ACF and PACF plots in tine-series
(h) What is the fundamental diflerence analysis?
between a P-P plot and a Q-Q plot?
(2
END