UNIVERSITY OF BOTSWANA

DEPARTMENT OF STATISTICS
STA 114: BUSINESS STATISTICS
END OF THE SEMESTER II FINAL EXAMINATION MAY 2015
Answer all Questions in Section A and Any Two Questions in Section B.
DURATION: 2 hrs

Section A

Answer ALL questions in this section.

Question A1

Determine whether the following statements are TRUE or FALSE

a) Sample median is called a parameter.

b) Weighted moving averages and exponential smoothing are both forecasting methods in time
series.
c) A binomially distributed random variable counts the number of events of interest occurring
within a fixed.

d) If a random variable X has Poisson distribution with mean E(X)=0.5, then its
Variance, Var(X)=2.5.
e) The Y intercept β0 ofY=β0+ β1 t represents the predicted value of Y when t = 0.
f) Sampling with replacement means that events are independent from one another.

g) px = ; x = 1, 2, 3is a valid probability mass function.

h) A continuous random variable can take any fixed value with probability zero.
i) If a random variable X is known to be normally distributed with known population mean µ and
X −µ
variance σ 2 then the new variable z= is distributed as standard normal.
σ
j) A seasonal index is an index which measures the average magnitude of seasonal effects in a
particular time series.

×10 marks=20 Marks]

[2×

Question A2

Answer the following questions by filling in the missing word or short sentence

a) The difference between the actual Y values and the predicted Y
values is known as.
b) A probability function is a _____ that assigns probabilities to the various values of x.
c) Using simple linear regression model it is possible to estimate the for a given time point.
d) distribution is used to model the inter occurrence time that elapses between the successive
occurrences of events.
e) The number of occurrence of some random phenomenon over a specified interval of time
orspace is described as a process.
f) The variance of a binomial distribution with  = 5 and  = 0.35 is.
1
 g) A normal distribution is about the median.
h) The mean value of andistribution is 1/λ.f ( x) = λe−λx
i) If Y=β0+ β1 t + error model is fitted to a given data set on (t , Y) then the estimated change
in average Y per unit change in t is.
j) If we roll a balanced die once, the probability of getting a 3 or 5 is.

×10=20 Marks]
[2×

SECTION B

ANSWER ANY TWO QUESTIONS FROM THIS SECTION.

QuestionB1

a) Table 1.1 contains the total aircraft movements (international and domestic) for Maun Airport from
2013 to the 3rd quarter of 2014. Figure 1.1 shows the time series plot for the data of Table 1.1.

Table 1.1: Total aircraft movements( for Maun Airport from 2013 to the 3rd quarter of 2014
Year Quarter Total aircraft movements (Yt)
2013 1 10519
2 13204
3 15304
4 13878
2014 1 9609
2 13940
3 14841
Source: Statistics Botswana. Botswana Transport and Infrastructure Statistics, No: 2014/03.
Stats Brief Quarter 3, 2014.

Figure 1.1: Number of aircraft movements at Maun Airport, Q1 2013 - Q3 2014

m
18000
o
16000
v
14000
e
12000
m
10000
e
8000 Number of movements (Y)
n
6000
t
4000
s
2000
0
1 2 3 4 5 6 7
Quarter

Note:1 to 4 represent quarters 1, 2, 3, 4 of the year 2013

5 to 7 represent quarters 1, 2, 3 of the year 2014
i) From Figure 1.1, what component(s) appears (appear) to be in the time series?
ii) Compute the 4-quarter centered moving averages for the series, writing your numbers to the
nearest whole number.

2
b) As part of an effort to forecast future sales, an operator of five independent gas stations recorded the
quarterly gasoline sales (in thousands of gallons) for the past 3 years. The data are shown in Table 1
below

Table 1.2: Quarterly gasoline sales for the past three years

Time Year Quarter Sales (‘000 gallons)

1 1 1 39
2 2 37
3 3 61
4 4 58
5 2 1 18
6 2 56
7 3 82
8 4 27
9 3 1 41
10 2 69
11 3 49
12 4 66

i) Draw a graph of the quarterly gasoline sales for the past 3 years.
ii) Apply exponential smoothing with  = 0.5 to the data in Table 1.2and plot the smoothed series
on the samegraph as the original sales data.
iii) What pattern can you see in the exponentially smoothed series compared to the original sales
series?

[2+7+5+15+1 = 30 Marks]

Question B2

a) Hospital records show that of patients suffering from a certain disease, 75% die of it. Six (6)
patients are selected at random. Suppose X denotes a random variable representing the number
of subjects who will recover from the disease.
i) What is the probability distribution of X?
ii) What is the probability that of the 6 randomly selected patients, more than 1 patient will
recover?
iii) What is the expected number of recoveries from this disease?

b) A life insurance salesman sells on the average 3 life insurance policies per week. Use Poisson's
law to calculate the probability that in a given week he will sell

i) At least one policy.

ii) 2 or more policies but less than 5 policies.

c) Based on the model in b) above, assuming that there are 5 working days per week, what is the
probability that in a given day he will sell three (3) or more policies?

[2+6+2+4+6+10 = 30 Marks]

3
Question B3

(a) The following table shows the quarterly sales for Toys International for the years 1996 through
2001. The sales are reported in millions of Pula. The table also presents the deseasonalized series
using the method of Ratio to moving average. Complete the table.

Year Quarter Yt Uncentered Centered Yt Seasonal SSI Yt

MA(4) *100 Indices SAS( *100
MA(4) St St SSI
)
1996 1 a - - - h 75.26 13.29
1996 2 9.20 16.15 - - 248.06 58.87 15.63
1996 3 20.00 16.90 c f 480.58 114.05 17.54
1996 4 25.40 16.90 16.90 150.30 639.69 151.81 16.73
1997 1 13.00 16.80 16.85 77.15 317.14 75.26 17.27
1997 2 9.20 17.25 17.03 54.04 248.06 58.87 15.63
1997 3 19.60 17.45 17.35 112.97 i 114.05 17.19
1997 4 27.20 18.15 17.80 152.81 639.69 151.81 17.92
1998 1 13.80 18.45 18.30 75.41 317.14 75.26 18.34
1998 2 12.00 18.70 18.58 64.60 248.06 58.87 m
1998 3 20.80 18.75 18.73 111.08 480.58 114.05 18.24
1998 4 28.20 18.50 18.63 151.41 639.69 k 18.58
1999 1 14.00 18.70 18.60 75.27 317.14 75.26 18.60
1999 2 b 19.15 d 58.12 248.06 58.87 18.68
1999 3 21.60 19.20 19.18 112.65 480.58 114.05 18.94
1999 4 30.00 19.30 19.25 155.84 639.69 151.81 n
2000 1 14.20 19.45 19.38 73.29 317.14 l 18.87
2000 2 11.40 19.20 19.33 58.99 248.06 58.87 19.36
2000 3 22.20 19.65 19.43 114.29 480.58 114.05 19.46
2000 4 29.00 19.90 e 146.65 639.69 151.81 o
2001 1 16.00 20.05 19.98 g 317.14 75.26 21.26
2001 2 12.40 20.25 20.15 61.54 248.06 58.87 21.06
2001 3 22.80 - - - 480.58 114.05 19.99
2001 4 29.80 - - - j 151.81 19.63

(b) Consider the following data, which shows shooting ranges for Botswana Army during training to
hit the target.

Actual (y t ) Forecast (yˆ t )

67.78 67.98
67.00 66.89
76.98 77.84
56.90 55.97
67.34 65.59
56.87 58.98
69.13 70.45
87.01 88.45

4
 (i) Obtain the Root mean square error (RMSE)
(ii) Obtain the Mean absolute forecast error (MAFE)
[15+8+7 = 30 Marks]

Question B4

Time series data on the number of ice-cream cones sold to a street vendor by the manufacturers of the
cones over the years 2000-2010 are observed as follows:

Year 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Number 35 42 48 52 58 55 50 65 51 75 74

a) Plot the time series data and suggest an appropriate model to fit the data.
b) Code the time such that ∑t=0 and fit the simple linear regression of Y on t.
c) Interpret all the estimated regression coefficients.
d) Obtain the trend values for 2000 and 2001
e) Assuming the above linear trend, forecast the trend values of the years 2014 and 2015.

[6+12+4+4+4 = 30 Marks]

Question B5

a) The promotional status of a major Metropolitan police force in the Greater Johannesburg
Suburb is given below. The police force consists of 1200 officers: 96 men and 240 women. Over
the last 2 years, 324 officers on the police force have been awarded promotion.

Officer Promoted Not promoted Total

Men 288 672 960
Women 36 204 240
Total 324 876 1200

Let ; M be an event that an officer is a man

W be an event that an officer is a woman
B be an event that an officer is promoted

Find the following:

i) P (M ∩ B )
ii ) P ( B / W )
iii ) P (W ∪ B )

b) A recent study of the hourly wages of maintenance crews for South African Airways showed that
their hourly salary (in RANDS) is normally distributed with mean 116.50, and standard deviation
12.00. If a crew member is selected at random, what is the probability that the crew member
earns:
i) Between 116.50 and 128.50 per hour.
ii) More than 140.50 per hour.
iii) Less than 100.00 per hour.

Given that P(Z ≤ −1.4) = 0.0808, P(0 ≤ Z ≤ 2) = 0.4772, P(Z ≤ 1) = 0.8413, when is a standard normal variable.

[4+5+4+6+6+5 =30 Marks]

5