MAASAI MARA UNIVERSITY
REGULAR UNIVERSITY EXAMINATIONS
2022/2023
SCHOOL OF BUSINESS AND ECONOMICS
BACHELOR’S OF SCIENCE IN ECONOMICS
AND STATISTICS
FOURTH YEAR SECOND SEMESTER
COURSE CODE: ECS 4204
COURSE TITLE: STATISTICAL INFERENCING
DATE: TIME:
INSTRUCTIONS:
Attempt Question one and any other Three Questions
1
Question One
a. For the following list of statements state whether the statements are TRUE
or FALSE giving arguments to support your decision.
i. When a hypothesis fails to be rejected at 95% confidence level it
can still be rejected at 99% confidence level. (2 marks)
ii. When we carry out a hypothesis test for mean difference at 95%
level of confidence and reject the null hypothesis. Then, when we
construct a 95% confidence interval for the mean then we expect
the confidence interval to contain zero. (2 marks)
iii. For any test involving the sample mean as a sample statistic used
in computing the test statistic it is mandatory that the distribution
of the sample should be approximately symmetric otherwise the
test cannot be done using sample mean. (2 marks)
b. The output below shows the results of a statistical test that was used to
test a hypothesis about the age of students samples from the university.
##
## Shapiro-Wilk normality test
##
## data: Age
## W = 0.9956, p-value = 0.8339
##
## One Sample t-test
##
## data: Age
## t = 8.3571, df = 199, p-value = 5.48e-15
## alternative hypothesis: true mean is greater than 20
## 95 percent confidence interval:
## 20.92419 Inf
## sample estimates:
## mean of x
## 21.15199
a. What was the sample size for the study? (1 mark)
b. State the null and alternative hypothesis for the test conducted.
(2 marks)
c. What was the assumption of the statistical test conducted in the extract
and what was the conclusion regarding the assumption based on the
output above at 95% level of confidence? (3 marks)
2
d. At 99% level of confidence, what was the conclusion regarding the null
and alternative hypothesis stated in (ii). (2 marks)
c. A sample of 25 students had an average height of 72 and average weight of
165. At 95% level of confidence, test the hypothesis that this sample was
taken from a bivariate normal population with average height of 71 and
20 100
average weight of 172 with a covariance matrix .
100 1000
(6 marks)
d. Discuss how the violation of each of the following assumptions by a linear
regression model affect an estimated regression model. (5 marks)
i. Normality
ii. Homoscedasticity
iii. Autocorrelation
iv. Multicollinearity
v. Model Specification
Question Two
From time to time, unknown to its employees, the research department at post
bank observes various employees for work productivity. Recently, this
department wanted to check whether the four tellers at a branch of this bank
serve, on average, the same number of customers per hour. The research
manager observed each of the four tellers for a certain number of hours. The
following table gives the number of customers served by the four tellers during
each of the observed hours.
Teller A Teller B Teller C Teller D
19 14 11 24
21 16 14 19
26 14 21 21
24 13 13 26
18 17 16 20
13 18
a. At the 5% significance level, test the null hypothesis that the mean
number of customers served per hour by each of these four tellers is the
same. (6 marks)
3
b. State the assumptions that you made when carrying out the test in part
(a) above. (2 marks)
c. How would you diagnose each of the assumptions made in part (b) in r
package? (2 marks)
d. State the type of error you might have committed in part (a) justifying
reasons for your conclusion. (2 marks)
e. If the company wished to compare the performance of Teller A and B
which parametric test would they use and what underlying assumptions
must be satisfied by the sample data for the test to be done? (3 marks)
Question Three
The manager of a certain sugar company wants to determine the most efficient
type of machine to use in production at the company. The manager believe that
the efficiency of the machine is measured by the amount of bags of sugar
produced per day. The company has 3 types of machines that it uses in
producing sugar. For a sample of days, the manager compared the number of
bags of sugar produced and the results were as illustrated below;
##
## Shapiro-Wilk normality test
##
## data: MachineA
## W = 0.55409, p-value = 7.448e-13
##
## Shapiro-Wilk normality test
##
## data: MachineB
## W = 0.80196, p-value = 5.475e-07
##
## Shapiro-Wilk normality test
##
## data: MachineC
## W = 0.74234, p-value = 1.777e-08
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 2 4.4395 0.01319 *
## 171
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Df Sum Sq Mean Sq F value Pr(>F)
## Machine 2 7.2 3.602 4.83 0.00911 **
## Residuals 171 127.5 0.746
4
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Kruskal-Wallis rank sum test
##
## data: Bags by Machine
## Kruskal-Wallis chi-squared = 11.605, df = 2, p-value = 0.00302
## Tukey multiple comparisons of means
## 99% family-wise confidence level
## factor levels have been ordered
##
## Fit: aov(formula = Bags ~ Machine, data = Production)
##
## $Machine
## diff lwr upr p adj
## Machine C-Machine B 0.2322470 -0.25856147 0.7230555 0.3445977
## Machine A-Machine B 0.4928531 0.02254435 0.9631618 0.0064875
## Machine A-Machine C 0.2606061 -0.20493589 0.7261480 0.2264870
a. State the null and alternative hypothesis for the study. (2 marks)
b. State the parametric test required to test the hypothesis in (a). (1 mark)
c. State two assumptions required in order to carry out the parametric test in
(b). (2 marks)
d. Based on the output given was the assumption in (c) satisfied, justify your
reasoning. (3 marks)
e. Given the conclusion in (d) what course of action was taken by the analyst
next. (1 mark)
f. Based on the findings in the output what conclusions could be on problem
in the study. Justify your conclusion. (3 marks)
g. Based on the conclusion made in (f) which type of error might have been
committed in (f). (2 marks)
h. Which machine(s) should the company invest its money in? (1 mark)
5
Question Four
36.09
Let Y y1 , y 2 , y 3 be a random vector with mean vector Y 25.55 and
34.09
65.09 33.65 47.59
covariance matrix S 33.65 46.07 28.95 . If the random variables Z and W is
47.59 28.95 60.69
formed through linear combination of Y such that Z 3 y 1 2 y 2 4 y 3 and
W y1 3 y 2 y 3 . Determine;
a. Z. (2 marks)
b. W. (2 marks)
c. var Z . (3 marks)
d. var W . (3 marks)
e. Correlation between Z and W . (3 marks)
f. Correlation between y 2 and y 3 . (2 marks)
Question Five
Two Psychological tests were given to 6 Males and 6 Females. The data record
was as illustrated in the table below.
Y1 = Pictorial Inconsistencies
Y2 = Paper form board
Male Female
Y1 Y2 X1 X2
25 27 13 14
27 25 14 12
25 24 12 19
28 26 17 17
27 20 12 15
29 29 16 14
Using the Hoteling’s T2 test, test the hypothesis that the psychological test
scores for Males is significantly different to that of Females at 95% confidence
level. (15 marks)
