2018 ABA 3692

[BUSINESS STATISTICS B]

[ABA3692]

Questions and answers

CENTRE FOR OPEN, DISTANCE AND ELEARNING

Open your mind

1 Compiled by Mr Reino Ihemba

 2018 ABA 3692

UNIT 1 SAMPLING DISTRIBUTIONS

Note 𝛿𝑥̅ is called standard error of mean or Standard deviation of the sampling distribution of
𝑥̅ .

Finite population: Infinite population

𝑁−𝑛 𝜎 𝜎
𝜎𝑥̅ = √𝑁−1 ( 𝑛) 𝜎𝑥̅ =
√ √𝑛

𝑁−𝑛
Note: The finite population correction factor√𝑁−1 , which is required when the population is
𝒏
finite and the sample size is more than 5% of the population size, i.e. > 𝟎. 𝟎𝟓. else the
𝑵
𝒏
correction factors is not necessary if ≤ 𝟎. 𝟎𝟓, were N stands for population size, while n is
𝑵
the sample size

Note the smaller the sample size the bigger the standard error.

 For Larger sample n ≥ 30

𝑥̅ −𝜇𝑜
Us Normal z--distribution) were Test statistic= Zc = 𝜎 the calculated Z value when
√𝑛
population variance 𝝈𝟐 is known else use t-distribution even if n ≥ 30 (but Using Z fall large
sample is still be fine because the larger sample tend to follow normal distribution).
Critical Statistic 𝑍𝛼 𝑎𝑛𝑑 𝑍𝛼 is the value to be located using the the Z table this is a value of Z
2
at p = 0.5 – 𝛼

 For Smaller sample size n < 30

𝑥̅ −𝜇𝑜
Use t-distribution were Test statistic = tc = 𝜎 to calculated t value when population
√𝑛

variance 𝝈𝟐 is unknown else use Z even if n < 30 (but Using t for all small samples is still fine).
Unknown/ not known variance (means when you only have a sample variance (s2) )

Critical Statistic 𝑡𝛼 or 𝑡𝛼 is the value to be located using the the t-table this is a value of is
2
fount at at 𝑡𝛼, (𝑛−1)𝑑𝑓 .
Parameters Population Sample
Mean 𝝁 ̅
𝒙
Variance 𝝈𝟐 s2
Standard deviation 𝝈 s

2 Compiled by Mr Reino Ihemba

 2018 ABA 3692

1. A population has mean of 300 and a standard deviation of 75. A simple random sample of size
150 will be taken and the sample mean will be used to estimate the population mean.

a. Is this a finite or infinite case

b. What is the standard error of mean?

Solution

a. This is an infinite case because the population size is not indicated only the sample size of 75 is
indicated
𝜎
b. Standard error of mean = 𝜎𝑥̅ = and we do not need the correction factor
√𝑛
𝜎 75
. 𝜎𝑥̅ = = = 6.12
√𝑛 √150

2. To estimate the mean duration of employment in the XPZ Company, a population of 3000 employees,
a simple random sample of 1000 employees is selected with a standard dev of 10.
a. a. Is this a finite or infinite case
b. What is the standard error of mean?
c. What is the standard error of mean if the sample was 50?
d. What is the standard error of mean if the sample was 50 using correction factor?
e. Comment on the answers of c and d

Solution

a. This is an finite case because the population size is indicated

𝑛 1000 𝑛
b. Standard error of mean = 𝜎𝑥̅ since . = = 0.333 means > 0.05. than
𝑁 3000 𝑁
we need the correction factor
𝑁−𝑛 𝜎 3000−1000 10 2000 10
. 𝜎𝑥̅ = √ ( 𝑛) = √ ( ) = √2999 ( ) = 0.26
𝑁−1 √ 3000−1 √1000 √1000

𝑛 50 𝑛
c. Standard error of mean = 𝜎𝑥̅ since . = = 0.017 means ≤ 0.05. than
𝑁 3000 𝑁
we do not need the correction factor
𝜎 10
. 𝜎𝑥̅ = ( ) = ( ) = 1.41
√𝑛 √50

𝑛 50 𝑛
d. Standard error of mean = 𝜎𝑥̅ since . = = 0.017 means ≤ 0.05. than
𝑁 3000 𝑁
we do not need the correction factor but applying the the correction factor
𝑁−𝑛 𝜎 3000−50 10 2950 10
. 𝜎𝑥̅ = √ ( ) = √ 3000−1 ( ) = √2999 ( ) = 1.40
𝑁−1 √𝑛 √50 √50

3 Compiled by Mr Reino Ihemba

 2018 ABA 3692

e. There seem to be no significant difference between the standard error of mean in c and in d. That
𝑛
is why answer in c because = 0.017 and ≤ 0.05 that is why there is no need for
𝑁
𝑛
correction factor if ≤ 0.05.
𝑁

3. Assume that the following random sample of 12 random sample of mean average scores
received by participants of Ms Namibia Pageant as rated judges are normal population
9,11,10; 8; 12; 14; 13; 11; 7; 8. 7,12

a. Compute the point estimate of the population mean

b. Compute the point estimate of the population standard deviation
c. Assume that the point estimate of mean and standard deviation are equal to the population mean
and standard deviation. Compute the standard error of mean

d. Assume that the point estimate of mean and standard deviation are equal to the population mean
and standard deviation. Compute the probability that a pageant receives a mean score is more
than 11 draw the distribution area.

Solution
Sum 𝑥̅
𝑥𝑖 9 11 10 8 12 14 13 11 7 8 7 12 122 10.17
- - -
-1.17 0.83 1.83 3.83 2.83 0.83 -3.17 -3.17 1.83 0
(𝑥𝑖 -𝑥̅ ) 0.17 2.17 2.17
(𝑥𝑖 -𝑥̅ )2 1.37 0.69 0.03 4.71 3.35 14.67 8.01 0.69 10.05 4.71 10.05 3.35 61.68

∑𝑥𝑖 61.68
(a) Point estimate of population mean = (𝑥̅ ) = 𝑛
= 12
= 10.17

∑(𝑥𝑖−𝑥
̅̅̅ )2 61.68
(b) Point estimate of population standard deviation = 𝑠 = √ 𝑛−1
= √ 12
= 𝟐. 𝟑𝟕

𝛿 2.37
(c) Standard error of mean = 𝛿𝑥̅ = = = 𝟎. 𝟔𝟖𝟒
√𝑛 √12
𝑥̅ −𝜇 11−10.17 0.83
(d) For 𝑥̅ = 11 scores, → 𝑧 = = = = 1.21 The probability at Z = 1.21 give p =
𝜎𝑥̅ 𝟎.𝟔𝟖𝟒 𝟎.𝟔𝟖𝟒
0.3869

Hence P ( 𝜇 > 11) = 0.5 − 0.3869 = 𝟎. 𝟏𝟏𝟑𝟏.

4 Compiled by Mr Reino Ihemba

 2018 ABA 3692

4. Employees at Kambwa Trading spend an average of 4 weeks job leave during the 2017
operation. Assume that 4 weeks is the population with a population standard deviation
of 1.581. Using a sample of 27 employees.

a. Determine the standard error of the mean.

b. Show the sampling distribution of the mean
c. What is the probability that staff mean leave falls between 5.5 to 6.5 weeks?
d. Draw shape of the sampling distribution of x in ( c)
e. What is the probability that mean leave is at least 7 weeks
f. What is the probability that mean leave is less or equal 2.5 weeks
g. Draw shape of the sampling distribution in (f)

Solution

a. Standard error of the mean. [4]

𝛿 1.581
𝛿𝑥̅ = ; 𝛿𝑥̅ = = 0.304
√𝑛 √27

b. Show the sampling distribution of the mean [3]

1.551
𝛿𝑥̅ = = 0.304
√27

c. P (5.5 ≤ 𝜇 ≤6.5) weeks (Note although the sample is less than 30 which means we should [5]
have used (t- distribution) but we cannot use t- distribution because variance 𝝈𝟐 is
known we and there is no 𝜶 (𝒂𝒍𝒑𝒉𝒂) or confidence level given in the question. Therefore
we use Z- distribution.
𝑥̅ −𝜇 5.5−4 1.5
For 𝑥̅ = 6.5 weeks, → 𝑧 = 𝜎𝑥̅
= 0.304
= 0.304 = 4.93
𝑥̅ −𝜇 6.5−4 2.5
and for 𝑥̅ = 9.5 𝑤𝑒𝑒𝑘𝑠 → 𝑧 = 𝜎𝑥̅
= 0.304 = 0.304 = 8.23
The Z table goes up 3.09, therefore, the highest probability at both 4.93 and 8.23 = 0.4990,
Therefore, the probability is 0.4990 - 0.4990 = 0.00

d. Draw the sampling distribution of 𝑥̅ in ( c) [3]

5 Compiled by Mr Reino Ihemba

 2018 ABA 3692

e. What is the probability that mean leave is at least 7 weeks [5]

𝑥̅ −𝜇 7−4
→𝑧= = = 9.87, Note the Z- table goes only up 3.09, then we use the highest probability
𝜎𝑥̅ 0.304
of 0.4990, Hence P ( 𝜇 ≥ 7) = 0.5 − 0.4990 = 𝟎. 𝟎𝟎𝟏.

f. What is the probability that mean leave is less or equal 2.5 weeks [5]
𝑥̅ −𝜇 2.5−4 −1.5
For 𝑥̅ = 6.5 weeks, → 𝑧 = 𝜎𝑥̅
= 0.304
= 0.304 = −𝟒. Note the Z- table goes only up 3.09,
then we use the highest probability of 0.4990, Hence P ( 𝜇 ≤ 2.5) = 0.5 − 0.4990 = 𝟎. 𝟎𝟎𝟏.

g. Draw the sampling distribution in (f) [3]

6 Compiled by Mr Reino Ihemba

 2018 ABA 3692

5. A total 350 Business statistics students out of a sample 600 who set for exams in
the first semester passed the course.

a. What is the point estimate of the proportion of the population of students who passed
the course?
b. What is the point estimate of the proportion standard deviation of students who
passed the course?
c. Assume that the point estimate in (a) is equal to the population proportion. What is
the probability that proportion of the population of students who passed the course is
more than 55%
d Draw shape of the sampling distribution of p¯ in (c)

Solution

a. Point estimate of p.

350
𝐸(𝑝̅) = 600 = 0.58

b. Point estimate of the proportion standard deviation

𝑝(1−𝑝) (0.58)(0.42) 0.2436
𝜎𝑝̅ = √ = √ =√ = 0.0201
𝑛 600 600

c. What is the probability that the proportion of dissatisfied employees is more than 0.35

𝑝̅ −𝑝 0.55−0.58 −0.03
.𝑧 = = = 0.0201 = −1.49
𝜎𝑝
̅ 0.0201

The probability at Z = 1.49 is P = 0.4319

Therefore, P ( 𝑝 < 0.33) = 0.5 + 0.4319 = 𝟎. 𝟗𝟑𝟏𝟗.

d Draw shape of the sampling distribution of p¯ in (c)

7 Compiled by Mr Reino Ihemba

 2018 ABA 3692

UNIT 2 INTERVAL ESTIMATION FOR POPULATION MEAN AND PROPORTIONS

𝜎
E = 𝑧𝛼 𝜎𝑥̅ = 𝑧𝛼 is called the Merging of error mean
2 2 √𝑛

𝑝(1−𝑝)
E = 𝑧𝛼 𝜎𝑝̅ = 𝑧𝛼 √ is called the Merging of error proportion
2 2 𝑛

2
(𝑧𝛼/2 ) 𝜎2
. 𝑛= sample size for the population mean
𝐸2

2
(𝑧𝛼 ) 𝑝(1−𝑝)
2
.𝑛 = = Sample size of the population proportion
𝐸2

𝝈
Confidence interval for the population mean = 𝐱̅ ∓ 𝒛𝜶
𝟐 √𝒏

̅(𝟏−𝒑
𝒑 ̅)
̅ ∓ 𝒛𝜶 √
Confidence interval for the population proportion = 𝒑
𝟐 𝒏

6. Rhino park hospital in Windhoek collected user profile data that showed that 30% of
the users were men. Assume that this percentage was based on a sample of 700 users.

a. What is the margin of error associated with the estimated proportion of users who are men at 95%
confidence level ?
b. Construct 95% confidence interval for the population proportion of the users who are men?
c. Construct 98% confidence interval for the population proportion of the users who are men?
d. With a .95 probability, what size of sample should be taken if the desired margin of error is 3%?

Solution
𝛼
a) α = 1 - 0.95 = 0.05 and = 0.025, n = 700, 𝑝̅ = 0.30
2
𝛼
Z = Z0.025 → 0.5-0.025= 0.475 in the table = 1.96, therefore
2
𝑝(1−𝑝) (0.3)(0.70)
∴ 𝐸 = 𝑧𝛼 √ = 1.96√ = 𝟎. 𝟎𝟑
2 𝑛 700

𝑝̅ (1−𝑝̅ )
b) The 95% confidence interval is 𝑝̅ ∓ 𝑧𝛼 √ = 𝑝̅ ∓ 0.03
2 𝑛

Since 𝑝̅ = 0.30 , 𝑝̅ ∓ .0.03

→ 0.30 − 0.03 ≤ 𝑝 ≤ 0.30 + 0.03 ∴. 𝟐𝟕 ≤ 𝒑 ≤ 𝟎. 𝟑𝟑

8 Compiled by Mr Reino Ihemba

 2018 ABA 3692

𝑝̅ (1−𝑝̅ )
c) The 98% confidence interval is = 𝑝̅ ∓ 𝑧𝛼 √
2 𝑛
𝛼
α = 1 - 0.98 = 0.02 and = 0.01
2
𝛼
Z 2 = Z0.01 → 0.5-0.01= 0.490 in the table = 2.33,
𝑝(1−𝑝) (0.30)(0.70)
∴ 𝐸 = 𝑧𝛼 √ = 2.33√ = 𝟎. 𝟎𝟒
2 𝑛 700

𝑝̅ (1−𝑝̅ )
= 𝑝̅ ∓ 𝑧𝛼 √ = 𝑝̅ ∓ .0.04 and since 𝑝̅ = 0.30 𝑝̅ ∓ .0.04
2 𝑛

→ 0.30 − 0.04 ≤ 𝑝 ≤ 0.30 + 0.04 ∴ 𝟎. 𝟐𝟔 ≤ 𝒑 ≤ 𝟎. 𝟑𝟒

2
(𝑧𝛼 ) 𝑝(1−𝑝)
2 (1.96)2 (0 .30)(0.70)
d) 𝑛 = = = 896.37 ≅ 𝟖𝟗𝟕.
𝐸2 0.032

7. A sample of 144 dogs showed an average weight of 16 kg. The Variance of the population is
known to be 1.96 kg.

a. Determine the standard error of mean.

b. Construct a 90% confidence interval for the mean of the population.
c. Construct a 70% confidence interval for the mean of the population.
d. Discuss why the answers in Parts a and b are different.
e. With a 0.95 probability, how large of a sample would have to be taken to provide a margin
of error of 0.16 kg

Solution
𝝈 𝟏.𝟒
a. Standard error of mean = 𝝈𝒙̅ = = = 0.12 kg
√𝒏 √𝟏𝟒𝟒

b. 99% confidence interval for for the mean of the population

𝜶
Note, Confidence intervals = 𝒙̅ ∓ 𝒛𝜶/𝟐 𝝈𝒙̅ and α = 1 - 0.9 = 0.01, = 0.005, n
𝟐
= 144 and 𝝈𝒙̅ = 0.12 kg
𝛼
Z 2 = Z0.005 → P= 0.5-0.005= 0.495 in the table lies between 0.4949 and 0.4951= hence
𝛼 2.57+2.58
Z2 = = 2.575
2
Therefore, the 99% confidence 𝑥̅ ∓ 𝑧𝛼/2 𝜎𝑥̅ = 16 ± (2.575)(0.12) = 16 ± 0.3
= 15.7kg ≤ 𝝁 ≤ 16.3 kg

c. 70% confidence interval for for the mean of the population

𝛼 𝝈 𝟏.𝟒
Note α = 1 - 0.70 = 0.30, = 0.15, n = 144 and 𝝈𝒙̅ = = = 0.12 kg
2 √𝒏 √𝟏𝟒𝟒

9 Compiled by Mr Reino Ihemba

 2018 ABA 3692
𝛼
Z 2 = Z0.15 → P= 0.5-0.15= 0.35 in the table Zα lies between 0.3485 and 0.3508
𝛼 1.03+1.04
hence Z 2 = = 1.035
2
Therefore, the 70% confidence = 16 ± (1.035)(0.12) = 16 ± 0.1242
= 15.9 kg ≤ 𝝁 ≤ 16.1kg

d. Why the answers in Parts (a) and (b) are different

Because the higher the confidence percentage level the wider the confidence interval.

e. With a 0.95 probability, the a sample size to provide a margin of error of 0.16 kg?
𝛼 𝛼
Not at P =0.95, α = 1 - 0.95 = 0.05 and 2 = 0.025, n = 144 and Z 2 = Z0.025 → 0.5-
0.025= 0.475 in the table = 1.96.

2
(𝑧𝛼/2 ) 𝜎2 (1.96)2 (1.4)2
Sample size = 𝑛 = = = 294.12 ≅ 𝟐𝟗𝟓 𝒅𝒐𝒈𝒔
𝐸2 0.162

10 Compiled by Mr Reino Ihemba

 2018 ABA 3692

8. A restaurant wanted to determine average amount of time customers spend in the

restaurant. A sample of 40 lunch customers was taken and average 45 minutes with
a standard deviation of 14 minutes was recorded.

a. Compute the standard error of the mean.

𝝈 𝟏𝟒
𝝈̅𝒙 = = = 2.2 mins
√𝒏 √𝟒𝟎

b. Construct a 75% confidence interval for the true average amount of time customers
spent in the restaurant. Because Variance 𝝈𝟐 is unknown we use t-distribution
even though n is more than 30

Confidence interval 𝒙̅ ∓ 𝒕𝜶/𝟐 𝝈𝒙̅

𝛼
Note α = 1 - 0.75 = 0.25 2 = 0.125, n = 40 and 𝜎𝑥̅ = 2.2 mins since t-table does not
have α =0.125 we use α =0.10 which is the closest to 0.125.
𝛼
t 2 (𝑛 − 1) = t 0.10 (40) = 1.2816
Therefore, 𝑥̅ ∓ 𝑡𝛼/2 𝜎𝑥̅ = 45 ± (1.2816)(2.2)
= 45 ± 2.82
= 42.18 ≤ 𝝁 ≤ 47.82 or (42.18, 47.82)

c. Construct a 90% confidence interval for the true average amount of time customers
spent in the restaurant.

Confidence interval 𝒙̅ ∓ 𝒕𝜶/𝟐 𝝈𝒙̅

𝛼
Note α = 1 - 0.90 = 0.1 2 = 0.0.05, n = 40 and 𝜎𝑥̅ = 2.2 mins
𝛼
Hence, t 2 (𝑛 − 1) = t 0.05 (40) = 1.6449
Therefore, 𝑥̅ ∓ 𝑡𝛼/2 𝜎𝑥̅ = 45 ± (1.6449)(2.2)
= 45 ± 3.62
= 41.38 ≤ 𝝁 ≤ 48.62 or (41.38, 48.62)

d Discuss why the answers in Parts (a) and (b) are different

The higher the confidence percentage levels the wider the confidence interval. In this case the
confidence interval at 90% is 8.86 minutes wider compare to 6.84 at 75% confidence interval.

e. With a .95 probability, how large of a sample would have to be taken to provide a
margin of error of 3 minutes or less?
𝛼 𝛼
Not at P =0.95, α = 1 - 0.95 = 0.05 and 2 = 0.025, and Z 2 = Z0.025 → 0.5-0.025=
0.475 in the table = 1.96.
2
(𝑧𝛼 ) 𝑠2
2 (1.96)2 (14)2
Hence the population size = 𝑛 = 𝐸2
= 32
= 83.66 ≅ 𝟖𝟒

11 Compiled by Mr Reino Ihemba

 2018 ABA 3692

UNIT 3 HYPOTHESIS TESTING ABOUT POPULATION MEANS

Note

 The Null hypothesis (H0)

o It represents the status quo to the party performing the sampling experiment.
o It is what is known (assumed to be true)
o The null hypothesis is accepted unless the data proves otherwise.

 The alternative or research hypothesis ((Ha)

o It is a claim contrary to status quo
o It is what the test is attempting to establish
o It will be accepted only if data provide convincing evidence of its truth.

 Critical value approach for larger samples takes the hypotheses test decision by
𝑥̅ −𝜇𝑜
comparing test statics = Zc = 𝜎 and critical statistic
√𝑛

𝑍𝛼 (𝟐 𝒔𝒊𝒅𝒆𝒅 𝒄𝒂𝒔𝒆)𝑜𝑟 𝑍𝛼 (𝒐𝒏𝒆 − 𝒔𝒊𝒅𝒆𝒅 𝒄𝒂𝒔𝒆) is the value to be located using the the
2
Z table this is a value of Z at p = 0.5 – 𝛼 . Note Z is used when population variance
𝝈𝟐 is known else use t-distribution even if n ≥ 30 (but Using Z fall large sample is still
be fine).

 Critical value approach for small samples (n < 30)takes the hypotheses test decision by
𝑥̅ −𝜇𝑜
comparing test statics = tc & critical statistic = tα, were tc = 𝜎 and
√𝑛

𝑡𝛼 (𝟐 𝒔𝒊𝒅𝒆𝒅 𝒄𝒂𝒔𝒆) 𝑜𝑟 𝑡𝛼 (𝟐 𝒔𝒊𝒅𝒆𝒅 𝒄𝒂𝒔𝒆) is the value to be located using the t-table at
2
𝑡𝛼,(𝑛−1)𝑑𝑓 . Note t- distribution is used when population variance 𝝈𝟐 is unknown else
use Z even if n < 30 (but Using t for all small samples is still be fine). Unknown/ not
known variance (means when you only have a sample variance (s2) )

 ρ value approach takes the hypotheses test decision by comparing p value at α against
the p value = 0.5- (probability at Zc in Z the table) for one sided test. Multiply p-value by
two for two sided case

 Let μo denote the specific numerical value being considered in the null and alternative
hypotheses. In general, a hypothesis test about the values of a population mean μ must
take one of the following forms.

Ho : μ ≥ μ o Ho : : μ ≤ μ o H0 : μ = μ 0

12 Compiled by Mr Reino Ihemba

 2018 ABA 3692

Ho : μ < Ho Ha : μ > μ o Ha : μ ≠ μ o

The choice of Ho and Ha is in many situations not so obvious and judgement is necessary to select
the proper form. However, as the preceding forms show, the equality part of the expression
(either ≥, ≤, or = always appears in the null hypothesisand < or > or ≠ μ0 appears in the
alternative hypothesis).

When formulate hypotheses about a population mean and a population proportion.

• We uses a Z test for the samples(n ≥30) and only when variance 𝝈𝟐 is known else use
t-distribution even if n ≥ 30
• We use T- test for samples (n < 30) to determine whether x̅ deviates significantly to
justify rejecting the null hypothesis

Steps for Hypothesis testing

1 State the null and alternative hypothesis

2 Specify alpha(𝛼) level and rejection region
3 Compute tests statistics and Critical Statistics / P-value
4 Compare tests statistics and Critical Statistics / P-value and alpha (𝛼)to the rejection
region
5 Make conclusions

13 Compiled by Mr Reino Ihemba

 2018 ABA 3692

9. It is known that the mean filing quantity for Chloride containers at Windhoek
Chemicals is 5 mg with the population standard deviation is 𝝈 = 𝟎. 𝟏𝟖 ; sample of 26
containers provided the mean of 4.92 mg.

a. State the null and alternative hypotheses.

b. Test null hypothesis at the 1% level of significance. Using the critical value approach
c. Test null hypothesis at the 1% level of significance. Using the p value approach
d. What do you conclude about the population mean in both (a) and (b) above?

Solution

a. 𝐻𝑜 : 𝜇 ≥ 5
𝐻𝑎 : 𝜇 < 5

b. Test null hypothesis at the 1% level of significance, using the critical value approach
We are using z- distribution because variance 𝝈𝟐 is known even though n is less than
30. Note This is a one sided lower tail test

At α = 0.01, we reject Ho if Z𝑐 < Z𝛼

𝑥̅ −𝜇𝑜 4.92−5 −0.08
Zc = 𝜎 = = 0.18 = 0.03 = -2.67 since Zc is negative we make Z α negative
√𝑛 √36

2.32+2.33
Also then since, Zα = Z0.01 = = - 2.325
2

Since Zc < Zα as -2.67 < −2.4851 ∴ we reject the Ho.

c. Test null hypothesis at the 1% level of significance , using the P- value approach

At α = 0.01 we reject Ho if P ≤ α = 0.01

𝑥̅ −𝜇𝑜 (𝑥̅ −5) 4.92−5 −0.08

Zc = 𝜎 𝜎 = 0.18 = 0.03 = -2.67 in the table Z= -2.67 → 0.4962 Therefore P value =
√𝑛 √𝑛 √36
0.5-0.4962 = 0.004

Since P < α as 0.004 < 0.01 ∴ we reject the Ho.

d. Using both the p value and critical value approach. We have rejected the H0, hence the population
mean is less than 5mg.

14 Compiled by Mr Reino Ihemba

 2018 ABA 3692

10. The Clicks cosmetics company believes it has a 35 percent share of the market in skin
care products in the country. This claim is questioned by a competitor, the Edgars
cosmetics saying Clicks cosmetics’ market share is less than as report. Edgars
consequently undertook a market research to survey females who use skin care
products. The market research sampled 400 females who used skin care products of
who 120 indicated that they used the Clicks cosmetic exclusively.

a. Compute the standard error of proportion of Clicks cosmetics market share

b. State the null and alternative hypotheses to be tested.
c. Test null hypothesis at the 5% level of significance. Using the critical-value approach
d. Conclude using the results obtained in ( c)

Solution

𝑝(1−𝑝) (0.35)(0.65) 0.2275

a. 𝜎𝑝̅ = √ = √ =√ = 0.0239
𝑛 400 400

b. Ho :𝑝 ≥ 0.35
Ha: 𝑝 < 0.35

120
c. Note α = 0.05 , 𝑝̅ = = 0.30, and this is a one sided lower tail test .
400
Reject Ho if Zc < Zα and since, Zα = Z0.05 → 0.5-0.05= 0.495 in the table Zα at
2.57+2.58
P=0.495, lies between 0.4949 and 0.4951 = hence Zα = = 2.575
2
𝑝̅ −𝑝 0.30−0.35 −0.05
𝑧= = = 0.0239 = −2.09 ∴ Since 𝑍𝑐 > 𝑍𝛼 𝑎𝑠 − 2.09 > −2.575, We
𝜎𝑝
̅ 0.0239

fail to reject the Ho

d. We conclude that the market share for Clicks cosmetics at least 35% hey have reported

11. Mean monthly basic salary for UNAM employees is 𝝁 = 𝟏𝟖, 𝟔𝟖𝟖 and NANTU claims
that mean salary is not what UNAM says. A sample of 400 UNAM employees shows a
sample mean salary of $16,860 and a standard deviation of $14,624?

a State the null and alternative hypotheses to be tested

b Test the null hypothesis at 0.05 level of significance using the critical value approach
c Test the null hypothesis at 0.3 level of significance using the p value approach

Solution

15 Compiled by Mr Reino Ihemba

 2018 ABA 3692

a. 𝐻𝑜 : 𝜇 = 18.688
𝐻𝑎 : 𝜇 ≠ 18.688

b. This is a two sided tail test since then

At α = 0.05 Reject Ho if 𝑡c < − 𝑡𝛼 or tc > 𝑡𝛼
2 2
𝑥̅ −𝜇𝑜 16.860−18.688
Since tc = 𝜎 = 14.624 = - 2.5, and 𝑡𝛼 = t0.025, 400 df = 1.96
√𝑛 √400 2

Therefore, 𝑡c < − 𝑡𝛼 as -2.5 < -1.96 we reject 𝐻𝑜

2
Then conclude that the mean annual income is not equal to N$18.688

c. At α = 0.05 level of significance , using the P- value approach

Reject Ho if P ≤ α = 0.05
𝑥̅ −𝜇𝑜 16.860−18.688 −1.828
Zc = 𝑠 = 14.624 = = -2.5
0.7312
√𝑛 √400

Using the table probability at Z = - 2.5 is 0.4798

Therefore P-value = 0.5-0.4962 = 2(0.0202) since this is a two tailed test we multiply
the value by two.
Since P < α as 0.0404 < 0.05 ,we reject 𝐻𝑜 , and conclude that the mean annual income
is not equal to N$18.688

16 Compiled by Mr Reino Ihemba

 2018 ABA 3692

UNIT 4 INTERVAL ESTIMATION AND HYPOTHESIS TESTING ABOUT THE

POPULATION VARIANCES

When Formulate hypotheses about a population variance we uses a Chi-square test statistics to
determine whether 𝜎 2 deviates significantly to justify rejecting the null hypothesis

Note the following in this Unit

(𝑛−1)𝑠2
 Chi-square test statistics = 𝜒 2 = 𝜎𝑜2
2
 One-sided 𝒍𝒐𝒘𝒆𝒓 𝒕𝒂𝒊𝒍 𝒕𝒆𝒔𝒕 uses 𝜒(1−𝛼),(𝑛−1)𝑑𝑓 as the Chi-square Critic statistics when
formulating your rejection region
2 2
i.e Reject Ho if 𝜒 2 < 𝜒(1−𝛼),(𝑛−1)𝑑𝑓 or simply Reject Ho if 𝜒 2 < 𝜒(1−𝛼)

2
 One-sided 𝒖𝒑𝒑𝒆𝒓 𝒕𝒂𝒊𝒍 𝒕𝒆𝒔𝒕 use 𝜒𝛼,(𝑛−1)𝑑𝑓 as the Chi-square Critic statistics when
formulating your rejection region
2
i.e Reject 𝐻𝑜 if 𝜒 2 > 𝜒𝛼,(𝑛−1)𝑑𝑓 or simply Reject 𝐻𝑜 𝜒 2 > 𝜒𝛼2

(𝑛−1)𝑠2 (𝑛−1)𝑠2
 Interval for Variance = 2 ≤ 𝜎2 ≤ ,
𝜒𝛼/2 𝜒2 𝛼
(1− )
2

(𝑛−1)𝑠2 (𝑛−1)𝑠2
 Interval for Standard deviation = √ 2 ≤ 𝜎2 ≤ ,
𝜒𝛼 𝜒2 𝛼
2 (1− )
2

Note the difference between the lower and upper bound denominators of the confidence interval
and the one-sided lower and upper tail Chi-square Critic statistics

17 Compiled by Mr Reino Ihemba

 2018 ABA 3692

12. Pupkewitz MadgeTech reported that the MadgeTech production variance of process is
less or equal to 25. A sample of 31 production process was tested and variance of 28.

a. Construct a 50% confidence interval for the population variance.

b. Construct a 95% confidence interval for the population standard deviation.
c. State the null and alternative hypotheses to be tested.
d. Test the null hypothesis is at the 5% level of significance.

e. What do you conclude about the population variance?

Solution
a. 50% confidence intervals for the variance. 𝑆𝑖𝑛𝑐𝑒 𝜎 2 = 25 since 𝑠 2 = 28 𝑎𝑛𝑑 𝑛 = 31
α = 0.5
(𝑛−1)𝑠2 (𝑛−1)𝑠2
2 ≤ 𝜎2 ≤ , and from the chi-square distribution table with 30 degrees of
𝜒𝛼/2 𝜒2 𝛼
(1− )
2

freedom 𝜒(2𝛼) = 𝜒(20.5) = 𝜒0.25,

2
30 𝑑𝑓 = 34.7997
2 2
2 2
𝜒(1− 𝛼 = 𝜒0.75, 30 𝑑𝑓 = 24.4776
)
2
(𝑛−1)𝑠2 (𝑛−1)𝑠2 (30)(28) (30)(28)
Therefore, 2 ≤ 𝜎2 ≤ , = .= ≤ 𝜎2 ≤
𝜒𝛼/2 𝜒2 𝛼 2
𝜒0.25 2
𝜒0.75
(1− )
2
840 840
= 34.7997 ≤ 𝜎 2 ≤ 24.4766 = 𝟐𝟒. 𝟏𝟒 ≤ 𝝈𝟐 ≤ 𝟑𝟒. 𝟑𝟐

b. 95% confidence intervals for the standard deviation. Hence α = 0.05

Note 𝜒(20.05) = 𝜒0.025,
2
30 𝑑𝑓 = 46.9792
2
2 2
𝜒(1− 𝛼 = 𝜒0.975, 30 𝑑𝑓 = 16.7907
)
2

Since standard dev = 𝜎 = √𝜎 2 hence, confidence intervals for the standard deviation is
(𝑛−1)𝑠2 (𝑛−1)𝑠2 (30)(28) (30)(28) 840 840
√ 2 ≤ 𝜎2 ≤ ,= √ ≤ 𝜎2 ≤ = √46.9792 ≤ 𝜎 2 ≤ 16.7907
𝜒𝛼 𝜒2 𝛼 2
𝜒0.025 2
𝜒0.975
(1− )
2 2

= √17.88 ≤ 𝜎 2 ≤ 50.03 = 𝟒. 𝟐𝟑 ≤ 𝝈 ≤ 𝟕. 𝟎𝟕

c. Null and alternative hypotheses to be tested, 𝑁𝑜𝑡𝑒 𝜎 2 = 25

Ho: 𝜎 2 ≤ 25
Ha: 𝜎 2 > 25

18 Compiled by Mr Reino Ihemba

 2018 ABA 3692

d. Test the company's claim at the 5%% level of significance.

2
Reject Ho if 𝜒 2 > 𝜒𝛼,(𝑛−1)𝑑𝑓 and note this is a one sided lower tail test.
(𝑛−1)𝑠2 (30)28 840
𝜒2 = = = = 33.6
𝜎𝑜2 25 25

And 𝜒𝛼2 (𝑛−1) 𝑑𝑓 = 𝜒0,05,

2
30 𝑑𝑓 = 43.77297 and
2
Since 𝜒 2 < 𝜒𝛼,(𝑛−1)𝑑𝑓 as 33.6 < 43.77297 18.4926, ∴ we do not reject the Ho or we
accept the Ho
e. We conclude that the production process variance is r 25 as reported by Pupkewitz
MadgeTech

13. Namibia Elite Provident Fund (NEPF) is known to have a variance of 35% in annual returns.
An Economic Research Society claims that the variance of NEPF returns is not true. A ample
of 12 annual returns was found to have a standard deviation 5%.

a. State the null and alternative hypotheses to be tested.

b. Test the Economic Research Society’s claims at the 10% level of significance
c. What do you conclude about the population variance?
d. Construct a 95% confidence interval for the population standard deviation.

Solution
a. Null and alternative hypotheses to be tested
𝐻𝑜 : 𝜇 = 35%
𝐻𝑎 : 𝜇 ≠ 35%

b. Test the claim at the 10% level of significance.

This is a two sided lower tail test at α = 0.1 we Reject Ho if 𝜒 2 < 𝜒 2 𝛼 or 𝜒 2 > 𝜒(2𝛼) .
(1− ) 2
2

Hence 𝜒2 𝛼 = 2
𝜒0.95, 11 𝑑𝑓 = 4.57481 and 𝜒 2𝛼 (𝑛 − 1) = 2
𝜒0.05, 11 𝑑𝑓 = 19.67514
(1− )(𝑛−1) ( )
2 2
(𝑛−1)𝑠2 (11)25 275
And 𝜒 2 = 𝜎𝑜2
= 35
= 35
= 7.86

Since 𝜒 2 > 𝜒 2 𝛼 or 𝜒 2 < 𝜒(2𝛼) .as 7.86> 4.57481, and 7.86 < 19.67514 ∴ we fail to reject
(1− ) 2
2

the Ho

c. We conclude that the annual returns is indeed 35%.

19 Compiled by Mr Reino Ihemba

 2018 ABA 3692

d. 95% confidence intervals for standard deviation is, 𝑆𝑖𝑛𝑐𝑒 𝑡ℎ𝑒𝑛 𝜎 2 = 35 since s = 7 then
𝑠 2 = 52 = 25 𝑎𝑛𝑑 𝑛 = 12 and α = 0.05
(𝑛−1)𝑠2 (𝑛−1)𝑠2
Population standard deviation =√ 2
𝜒𝛼/2
≤ 𝜎 ≤ √ 𝜒2 , and from the chi-square distribution
𝛼
(1− 2 )

table with 11 degrees of freedom 𝜒(2𝛼) = 𝜒(20.05) = 𝜒0.025,

2
11 𝑑𝑓 = 21.92005
2 2

𝜒2 𝛼
2
= 𝜒0.975, 11 𝑑𝑓 = 3.81575
(1− )
2

(11)(25) (11)(25) 275 275

→√ 2
𝜒0.025
≤𝜎≤√ 2
𝜒0.975
→ √21.92005 ≤ 𝜎 ≤ √3.81575 = 3.54≤ 𝝈 ≤ 𝟖. 𝟒𝟗

14. CODEL claims that the standard deviation of the study guide delivery time to
distance students less than 8 days after registration. A sample of 30 studded
was taken and a standard deviation 10 days delivery time was recorded.

a. State the null and alternative hypotheses to be tested.

b. Construct a 90% confidence interval for the population standard deviation.
c. Test the CODEL’s claim at the 10%% level of significance and state the decision rule
for the test.
d. Test the CODEL’s claim at the 5% level of significance and state the decision rule for
the test.

Solution

a. Null and alternative hypotheses to be tested

Ho: 𝜎 2 ≥ 64
Ha: 𝜎 2 < 64

b. Test the claim at the 10% level of significance.

2
Reject Ho if 𝜒 2 < 𝜒(1−𝛼) , this is a one sided lower tail test with α = 0.1
2 2
hence 𝜒1−𝛼 = 𝜒0,90 and from the chi-square distribution table with 29 degrees of
freedom.
2 (𝑛−1)𝑠2 (29)100 2900
𝜒0,90, 29 𝑑𝑓 = 19.7677, and 𝜒 2 = = = = 45.31
𝜎𝑜2 64 64
2
Since 𝜒 2 > 𝜒(1−𝛼) as 45.31> 19.7677, ∴ we do not reject the Ho or we accept the Ho
the variance of delivery time is higher than 64 days

20 Compiled by Mr Reino Ihemba

 2018 ABA 3692

c. Test the claim at the 5%% level of significance.

2
At α = 0.05 Reject Ho if 𝜒 2 < 𝜒(1−𝛼) , this is a one sided lower tail test
2 2
hence 𝜒1−𝛼 = 𝜒0,95 and from the chi-square distribution table with 29 degrees of
2 (𝑛−1)𝑠2 (29)100 2900
freedom𝜒0,95, 29 𝑑𝑓 = 17.7083, and 𝜒 2 = = = = 45.31
𝜎𝑜2 64 64
2
Since 𝜒 2 > 𝜒(1−𝛼) as 45.31> 17.7083, ∴ we do not reject the Ho or we accept the Ho
the variance of delivery time is higher than 64 days

d. 90% confidence intervals for standard deviation is, 𝑆𝑖𝑛𝑐𝑒 𝜎 = 8 𝑡ℎ𝑒𝑛 𝜎 2 = 82 = 64

since s = 10 then 𝑠 2 = 102 = 100 𝑎𝑛𝑑 𝑛 = 30 and α = 0.1

(𝑛−1)𝑠2 (𝑛−1)𝑠2
Population standard deviation =√ 2 ≤ 𝜎 ≤ √ 𝜒2 , and from the chi-square
𝜒𝛼/2 𝛼
(1− )
2

distribution table with 29 degrees of freedom 𝜒(2𝛼) = 𝜒(20.1) = 𝜒0.05,

2
26 𝑑𝑓 = 42.5569
2 2

2 2
𝜒(1− 𝛼 = 𝜒0.95, 26 𝑑𝑓 = 17.7083
)
2

(29)(100) (29)(100) 2900 2900

→√ 2 ≤𝜎≤√ 2 → √42.5569 ≤ 𝜎 ≤ √17.7083
𝜒0.025 𝜒0.975

= √68.14 ≤ 𝜎 ≤ √163.77 = 𝟖. 𝟐𝟓 ≤ 𝝈 ≤ 𝟏𝟐. 𝟖

21 Compiled by Mr Reino Ihemba

 2018 ABA 3692

UNIT 5 LEANER REGRESSION

15. The size of space allocated to groups or individual exhibitors at Katima trade fair was
allocated as follows: the space size per member is based on a number of members in a
group

Group # of members x Space size

A 2 25
B 3 25
C 5 20
D 1 30
E 8 16
a) Develop an estimated regression equation.
b) Using the regression equation in part(a) Compute 𝑆𝑆𝐸, 𝑆𝑆𝑇, and 𝑆𝑆𝑅.
c) Compute the coefficient of determination.
d) Interpret the results in c
e) Compute the sample correlation coefficient.
f) Interpret the results in d

Solution
x y x2 xy ̂ = 𝟑𝟎. 𝟑𝟑 − 𝟏. 𝟖𝟕𝟕𝒙.
𝒚 ̂ 𝒊 )𝟐
(𝒚𝒊 − 𝒚 ̅𝒊 )𝟐
(𝒚𝒊 − 𝒚 (𝒚 ̅ 𝒊 )𝟐
̂𝒊 − 𝒚
2 25 4 50 26.6 2.5 3.2 11.4
3 25 9 75 24.7 0.1 3.2 2.2
5 20 25 100 20.9 0.9 10.2 5.1
1 30 1 30 28.5 2.4 46.2 27.6
8 16 64 128 15.3 0.5 51.8 62.2
19 116 103 383 6.3 114.8 108.5
𝑥̅ 3.8 𝑺𝑺𝑬 𝑺𝑺𝑻 𝑺𝑺𝑹
𝑦̅ 23.2

a) 𝑦̂ = 𝑏0 + 𝑏1 𝑥 ,

(∑ 𝑥𝑖 ∑ 𝑦𝑖 ) (19)(116)
∑ 𝑥𝑖 𝑦 𝑖 − 383− 383−440.8 102.8
𝑏1 = 𝑛
2 = 5
(19)2
= = = −1.877
2 (∑ 𝑥𝑖 ) 103− 103−72.2 30.8
∑ 𝑥𝑖 − 5
𝑛

𝑏0 = 𝑦̅ − 𝑏1 𝑥̅ = 23.2 − (−1.877)(3,8) = 30.33

̂ = 𝟑𝟎. 𝟑𝟑 − 𝟏𝟖𝟕𝟕𝒙.
∴𝒚

22 Compiled by Mr Reino Ihemba

 2018 ABA 3692

b) 𝑆𝑆𝐸 = ∑(𝑦𝑖 − 𝑦̂𝑖 )2 = 6.33, 𝑆𝑆𝑇 = ∑(𝑦𝑖 − 𝑦̅𝑖 )2 = 𝟏𝟏𝟒. 𝟖 and 𝑆𝑆𝑅 = ∑(𝑦̂𝑖 − 𝑦̅𝑖 )2 =
𝟏𝟎𝟖. 𝟓
𝑆𝑆𝑅 108.86
c) 𝑟 2 = = = .948
𝑆𝑆𝑇 114.8

d) The coefficient of determination of 0.948 means that 94.8% of the variability in the data
is explained by using the estimated regression equation (model) and only 5.2% which is
not captured by these model.

e) 𝑟 = (𝑠𝑖𝑔𝑛 𝑜𝑓 𝑏1 )√𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑑𝑒𝑡𝑒𝑟𝑚𝑖𝑛𝑎𝑡𝑖𝑜𝑛 . = (𝑠𝑖𝑔𝑛 𝑜𝑓 𝑏1 )√𝑟 2 . Therefore

= (−)√. 948 = −. 𝟗𝟕𝟐

f) Since the value of r is close to -1, this means there is a very strong negative leaner
relationship number of members (x) and space size (y).

16. Below you are given information on a woman's age and her annual expenditure on
purchase of books.

Woman Age (x) Expenditure ($) (y)

A 18 210
B 22 180
C 21 220
D 28 280

a. Which variable is the dependent variable? Which is the independent variable?

b. Determine the least squares estimated regression line.
c. Predict annual book purchase expenditure of a woman of 30 years.

d. Calculate the coefficient of determination.

e Calculate the sample correlation coefficient between age of a woman and annual book
purchase expenditure.

f. Interpret the value you obtain

23 Compiled by Mr Reino Ihemba

 2018 ABA 3692

Solution

x y x2 xy ̂ = 𝟓𝟒. 𝟕𝟒 + 𝟕. 𝟓𝟒𝒙.
𝒚 ̂𝒊 )𝟐
(𝒚𝒊 − 𝒚 ̅𝒊 )𝟐
(𝒚𝒊 − 𝒚 (𝒚 ̅𝒊 )𝟐
̂𝒊 − 𝒚
18 210 324 3780 190.46 382.2025 156.25 1027.20
22 180 484 3960 220.62 1649.1721 1806.25 3.57
21 220 441 4620 213.08 48.0249 6.25 88.92
28 280 784 7840 265.86 200.2225 3306.25 1879.22
89 890 2033 20200 2279.62 5275 2998.92
𝑥̅ 22.25 SSE SST SSR
𝑦̅ 222.5

a. y the expenditure is the depended variable which depends on x the age of a woman
which is the independent variable.

b. The least squares estimated regression line.

(∑ 𝑥𝑖 ∑ 𝑦𝑖 ) (89)(890)
∑ 𝑥𝑖 𝑦 𝑖 − 20200−
̂ = 𝒃𝟎 + 𝒃𝟏 𝒙 , since 𝑏1 =
𝒚 𝑛
2 = 4
(89)2
= 7.54 and
2 (∑ 𝑥𝑖 ) 2033−
∑ 𝑥𝑖 − 4
𝑛

𝑏0 = 𝑦̅ − 𝑏1 𝑥̅ = 222.5 − (7.54)(22.25) = 54.74

̂ = 𝟓𝟒. 𝟕𝟒 + 𝟕. 𝟓𝟒𝒙.
∴𝒚

c. Annual book purchase expenditure of a woman of 30 years .

For x = 1 the income will be 𝑦̂ = 54.74 + 7.54𝑥. = 54.74 + 7.54(30) = N$ 280.94

d. Coefficient of determination.
𝑆𝑆𝑅
= 𝑟 2 = 𝑆𝑆𝑇 , note ̂𝒊 )𝟐 = 2279.62, 𝑺𝑺𝑻 = ∑(𝒚𝒊 − 𝒚
since 𝑺𝑺𝑬 = ∑(𝒚𝒊 − 𝒚 ̅𝒊 )𝟐 = 𝟓𝟐𝟕𝟓 and
𝑆𝑆𝑅 2998.92
𝑆𝑆𝑅 = ∑(𝑦̂𝑖 − 𝑦̅𝑖 )2 = 𝟐𝟗𝟗𝟖. 𝟗𝟐 and hence = 𝑟 2 = 𝑆𝑆𝑇 = = 𝟎. 𝟓𝟕
5275

e. Sample correlation coefficient 𝑟 = (𝑠𝑖𝑔𝑛 𝑜𝑓 𝑏1 )√𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑑𝑒𝑡𝑒𝑟𝑚𝑖𝑛𝑎𝑡𝑖𝑜𝑛

= (𝑠𝑖𝑔𝑛 𝑜𝑓 𝑏1 )√𝑟 2 . = (+)√0.57 = 𝟎. 𝟕𝟓.

f. Since the value of r is close to 1, this means there is a very strong positive leaner relationship between age
(x) of a woman and annual book purchase expenditure (y).
.

24 Compiled by Mr Reino Ihemba

 2018 ABA 3692

UNIT 6 INDEX NUMBERS

17. An Old Mutual client’s portfolio of shares and their prices in 2012 and 2017 are listed
in the table below. Use 2012 as base year.

2012 2017 2012 2017

Shares
Quantity Quantity Price Price
A 400 300 125c 275c
B 250 300 235c 185c
C 150 100 1 865c 2 515c

a. Compute the Laspeyres price and quantity indexes

b. Interpretation
c. Compute the Paasche price and quantity indexes
d. Interpretation

Price index Quantity index

𝑃 𝑄 𝑃 𝑄
a. Laspeyres 𝐼2017 = ∑ 1 0 (100) 𝐼2017 = ∑ 0 1 (100)
indexes. 𝑃𝑜 𝑄0 𝑃𝑜 𝑄0

(275)(400)+(185)(250)+(2515)(150) (125)(300)+(235)(300)+(1865)(100)
=(125)(400)+(235)(250)+(1865)(150) ∗ 100 =(125)(400)+(235)(250)+(1865)(150) ∗ 100

533500 294500
=388500 ∗ 100 = 137.32 =388500 ∗ 100 = 75.80

b. Interpretation The share price have increased by 37.32% Based on the current share prices the quanity
over the period from 2012 to 2017 that can be purchased has decreased 25 % over
the period from 2012 to 2017
𝑃 𝑄 𝑃 𝑄
c. Paasche 𝐼2017 = ∑ 𝑃1 𝑄1 (100) 𝐼2017 = ∑ 𝑃1 𝑄1(100)
indexes 𝑜 1 1 0

(275)(300)+(185)(300)+(2515)(100) (275)(300)+(185)(300)+(2515)(100)
=(125)(300)+(235)(300)+(1865)(100) ∗ 100 =(275)(400)+(185)(250)+(2515)(150) ∗ 100

389500 389500
=294500 ∗ 100 = 132.25 =533500 ∗ 100 = 73.01

d. Interpretation The share price have increased by 32.25% Based on the current share prices the quanity
over the period from 2012 to 2017 that can be purchased has decreased 26.9 %
over the period from 2012 to 2017

25 Compiled by Mr Reino Ihemba

 2018 ABA 3692

18. The flowing table contain prices and quantities of Bread, Milk, Sugar, Apple Juice and
Orange Juice products of Spar Shopping for the years 2008, 2011 and 2015.

Price Quantity
Product 2011 2014 2018 2011 2014 2018
Apple juice (500mil) 8 12 18 4 4 3
Orange Juice (500mil) 9 14 21 4 4 3
Butter (500g) 6 8 12 5 6 5
Milk(1litre) 7 15 22 5 3 3
sugar (2kg) 14 20 27 5 4 2

a. Compute the Paasche price and quantity indexes for 2015 using 2008 as the base year.
b. Provide interpretations of the Paasche price and quantity indexes in (a)
c. Compute the Paasche price and quantity indexes for 2015 using 2011 as the base year.
d. Provide interpretations of the Paasche price and quantity indexes in (c)
e. Compute the Laspeyres price and quantity indexes for 2015 using 2008 as the base year.
f. Provide interpretations of the Laspeyres price and quantity indexes in (e)
g. Compute the Laspeyres price and quantity indexes for 2015 using 2011 as the base year.
i. Provide interpretations of the Laspeyres price and quantity indexes in (g)
i. Comment on the results of Laspeyres comparison to Paasche indexes

Price index Quantity index

𝑃 𝑄 𝑃 𝑄
a. Paasche price 𝐼2018 = ∑ 𝑃1 𝑄1 (100) 𝐼2018 = ∑ 1 1(100)
𝑃1 𝑄0
and quantity 𝑜 1

indexes for 54+63+60+66+54 54+63+60+66+54

2015 using = 24+27+30+21+28 ∗ 100 = 72+84+60+110+135 ∗ 100
2008 as the
base year 297 297
=130 ∗ 100 = 228.46 =461 ∗ 100 = 64.43

b. Interpretation The product prices have increased by 128.46% over Based on the current prices the quanity that can be
the period from 2008 to 2015 purchased has decreased 35.57% over the period
from 2008 to 2015

26 Compiled by Mr Reino Ihemba

 2018 ABA 3692

𝑃 𝑄 𝑃 𝑄
c. Paasche price 𝐼2017 = ∑ 𝑃1 𝑄1 (100) 𝐼2017 = ∑ 𝑃1 𝑄1(100)
and quantity 𝑜 1 1 0

indexes for 54+63+60+66+54 54+63+60+66+54

2015 using = ∗ 100 = ∗ 100
36+42+40+45+40 72+84+72+66+108
2011 as the
base year 297 297
=203 ∗ 100 = 146.31 =402 ∗ 100 = 73.89

d. Interpretation The product prices have increased by 46.31% over the Based on the current prices the quanity that can be
period from 2011 to 2015 purchased has decreased 26.11 % over the period
from 2011 to 2015
𝑃 𝑄 𝑃 𝑄
e. Laspeyres 𝐼2018 = ∑ 𝑃1 𝑄0 (100) 𝐼2018 = ∑ 𝑃0 𝑄1 (100)
indexes 2015 𝑜 0 𝑜 0

using 2008 as 72+84+60+110+135 24+27+30+21+28

the base year = ∗ 100 =32+36+30+35+70 ∗ 100
32+36+30+35+70

461 130
=203 ∗ 100 = 227.09 =203 ∗ 100 = 64.04

f. Interpretation The product prices have increased by 127.09% over Based on the current prices the quanity that can be
the period from 2008 to 2015 purchased has decreased 35.96 % over the period
from 2008 to 2015

𝑃 𝑄 𝑃 𝑄
g. Laspeyres 𝐼2018 = ∑ 𝑃1 𝑄0 (100) 𝐼2018 = ∑ 𝑃0 𝑄1 (100)
indexes 2018 𝑜 0 𝑜 0

using 2008 as 72+84+72+66+108 36+42+40+45+40

the base year = 48+56+48+45+80 ∗ 100 = ∗ 100
48+56+48+45+80

402 203
=277 ∗ 100 = 145.13 = ∗ 100 = 73.29
277

h. Interpretation The product prices have increased by 45.13% over the Based on the current prices the quanity that can be
period from 2011 to 2015 purchased has decreased 26.71 % over the period
from 2011 to 2015

i. Comment on the results of Laspeyres comparison to Paasche indexes.

Both Laspeyres and Paasche indexes have shown the same trend in terms of price increase of the given period and
the reduction in quantise that can be purchased.

27 Compiled by Mr Reino Ihemba