You are on page 1of 16

Student Number: (enter on the line below)

Student Name: (enter on the line below)

HI6007
STATISTICS FOR BUSINESS DECISIONS
FINAL ASSESSMENT
TRIMESTER 1, 2022

Assessment Weight: 50 total marks

Instructions:
 All questions must be answered by using the answer boxes provided in
this paper.
 Completed answers must be submitted to Blackboard by the published
due date and time.

Please ensure you follow the submission instructions at the end of this paper.

Purpose:
This assessment consists of six (6) questions and is designed to assess your level of
knowledge of the key topics covered in this unit.

HI6007 Final Assessment T1 2022


HI6007 Final Assessment T1 2022
Question 1 (7 marks)
The data in the table below presents the hourly quantity of production for three lines of
production processes over the first 4 days in XYZ Company. Answer the questions based
on the Excel Output given below.

Day Process 1 Process 2 Process 3


1 33 33 28
ANOVA: 2 30 35 36 Single Factor
3 28 30 30 SUMMARY
Groups 4 Count Sum
29 Average
38 Variance
34
Process 1 4 120 30 4.66667
Process 2 4 136 34 11.3333
Process 3 4 128 32 13.3333
ANOVA
Source of Variation SS df MS F P value
Between Groups 32 ?2 ? 16 ? 1.6364 0.24766
Within Groups 88 ?9 ? 9.77776
Total 120 11

A. State the null and alternative hypothesis for single factor ANOVA.
(1 mark)
B. State the decision rule (α = 0.05). (1.5
marks)
C. Calculate the test statistic. (3
marks)

HI6007 Final Assessment T1 2022


D. Make a decision. (1.5
marks)
ANSWER: ** Answer box will enlarge as you type

A.
𝐍𝐮𝐥𝐥 𝐇𝐲𝐩𝐨𝐭𝐡𝐞𝐬𝐢𝐬, 𝐇𝟎 : 𝛍𝐏𝐫𝐨𝐜𝐞𝐬𝐬 𝟏 = 𝛍𝐏𝐫𝐨𝐜𝐞𝐬𝐬 𝟐 = 𝛍𝐏𝐫𝐨𝐜𝐞𝐬𝐬 𝟑
Alternative Hypothesis, Ha : Not all the process means are
equal
Where 𝛍𝐏𝐫𝐨𝐜𝐞𝐬𝐬 𝟏 = mean hourly quantity of production for Processes 1 over the
first 4 days in XYZ Company
Where 𝛍𝐏𝐫𝐨𝐜𝐞𝐬𝐬 𝟐 = mean hourly quantity of production for Processes 2 over the
first 4 days in XYZ Company
Where 𝛍𝐏𝐫𝐨𝐜𝐞𝐬𝐬 𝟑 = mean hourly quantity of production for Processes 3 over the
first 4 days in XYZ Company

B.

p-value Approach: Reject H0 if p-value ≤ 𝟎. 𝟎𝟓


Critical Value Approach: Reject H0 if F ≥ F0.05 = 4.26
where the value of F is based on an F distribution with k - 1 numerator d.f. and
nT - k denominator d.f.
k – 1 = 3 -1 = 2
nT – k = 12 – 3 = 9
where F0.05 = 4.26 is from the F distribution with 2 numerators degree of freedom
and 9 denominator degree of freedom

C.

HI6007 Final Assessment T1 2022


Test Statistic, F = MSTR/MSE
MSTR, mean square between treatments as the sample sizes are equal

𝒙
ന= (𝒙ഥ𝟏 + 𝒙
ഥ𝟐 + 𝒙ഥ𝟑 )/3 = (30 + 34 + 32)/3 = 32
SSTR = 4(30 -32)2 + 4(34 -32)2 + 4(32 -32)2 = 16 + 16 =32
MSTR = 32/3-1 = 32/2 =16

MSE, mean square error


SSE = 3(4.66667) + 3(11.3333) + 3(13.3333) = 87.99981
MSE = 87.99981/(12 -3) =87.99981/9 = 9.77776

Therefore Test Statistic , F = 16/9.77776 = 1.6364

D.

1) Based on P – value : With 2 numerator degree of freedom and 9 degree of


freedom denominator, the p-value is 0.05 for F = 4.2565 is 0.24766 calculated
through excel using formula = FDIST(1.6364,2,9)

As the p – value is greater than 0.05 we fail to reject H 0 and conclude that there is
insufficient evidence that the mean hourly quantity of production for 3 processes
over the first 4 days in company XYZ are not all the same.
2) Based on Critical value Approach
Reject H0 if F ≥ F0.05 = 4.26 where F0.05 = 4.26 is from the F distribution with 2
numerators degree of freedom and 9 denominator degree of freedom.

The F from test statistic is 1.6364 as F < 4.26 we fail to reject H o and conclude that
there is insufficient evidence that the mean hourly quantity of production for 3
processes over the first 4 days in company XYZ are not all the same.

Question 2 (7 marks)
Assume you have noted the following prices for books and the number of pages that each
book contains.
Book Price (in $) Pages
Y X
A 7.00 500

HI6007 Final Assessment T1 2022


B 7.50 700
C 9.00 750
D 6.50 590
E 7.50 540
F 7.00 650
G 4.50 480

A part of the output of a regression analysis of Y against X using Excel is given below:
SUMMARY
OUTPUT

Regression Statistics
Multiple R 0.75027
R Square 0.56290
Adjusted R
Square 0.475487
Standard Error 0.980614
Observations 7

ANOVA
  df SS MS F Significance F
6.19197
Regression 1 6.191972 2
0.96160
Residual 5 4.808028 6
Total 6 11      

Coefficient Standard
  s Error t Stat P-value
Intercept 1.04155 2.37717
Pages 0.00990 0.00390

HI6007 Final Assessment T1 2022


A. State the estimated regression line and interpret the slope coefficient. (1
mark)
B. What is the estimated total price when a book has 1000 pages? (1
mark)
C. What is the value of the coefficient of determination? Interpret it. (1
mark)
D. Test whether there is a significant relationship between price and pages at the 10%
significance level. Perform the test using the following six steps.
(4 marks)
(0.5 mark)

ANSWER:

A.
Y = 0.00990X + 1.04155
Price (in $) = 0.00990 * Pages + 1.04155
Slope Coefficient Interpretation : - For every additional page of the book, price will
increase by $0.0099, provided all else remain constant.

B.
Total price when a book has 1000 Pages
Price (in $) = 0.00990*1000 + 1.04155
= 10.94155 $

C.
The value of coefficient of determination is obtained from R square which is 56.29%
and therefore we can interpret that 56.29% variation in price of a book is explained
by the pages in a book. Remaining 43.71% is not explained by this model

D.

HI6007 Final Assessment T1 2022


Step 1. Statement of the hypothesis
(0.5 mark)
Ho: β1 = 0 which indicates no statistically significant relationship between number of pages in a book
to the price of a book
Ha: β1 ≠ 0 which indicates statistically significant relationship between number of pages in a book to
the price of a book
Step 2. Standardised test statistic
(0.5 mark)
𝐛𝟏
𝐭=
𝐬𝐛𝟏
Where b1 = slope of the sample regression line
And 𝐬𝐛𝟏 = standard error of the slope.
Step 3. Level of significance
(0.5 mark)
As per the question, level of significance = 10% or 0.1
Step 4. Decision Rule
(1 mark)
We Reject Ho if calculated test – statistic, tcal > crtitical test statistic tcrit
Tcrit at 0.1 level of significance and degree of freedom n-1 = 6-1 =5 is 2.015
Therefore the decision rule is Reject Ho if tcal > 2.015
Step 5. Calculation of test statistic
(1 mark)
Test statistic, t = 1.04155/2.37717 = 0.43815
Step 6. Conclusion
As the test statistic is less than t critical value which is 2.015 we fail to reject H 0 and
we conclude at 10% level of significance that no statistically significant relationship
between number of pages in a book to the price of a book exists.

Question 3 (11 marks)


A reporter for a student newspaper is writing an article on the cost of off-campus
housing.  A sample was taken of 10 one-bedroom units within a half-mile of the campus
and the rents paid.  The sample mean is $550 and the sample standard deviation is
$60.05.  We assume the population for one-bedroom units is normally distributed.

HI6007 Final Assessment T1 2022


Your task is to construct a 95% confidence interval for the average rent per month for the
population by addressing the following:
A. Parameter of interest (0.5 mark)
B. Point estimator (0.5 mark)
C. Sampling distribution of the point estimator (0.5
mark)
D. Specify the formula for the 95% confidence interval estimator for the parameter
(1 mark)
E. Perform the necessary calculations and write down the lower and upper limits of
the confidence interval (3
marks)
F. Interpret the calculated confidence interval (2
marks)

G. Briefly explain what would happen to the width of the interval in each case: (i) the
sample size increased, (ii) the sample standard deviation increased, and (iii) the
level of confidence increased (3
marks)

ANSWER:
A.

Population Mean which is the average rent per month for the population
95% of the sample means that can be observed are within + 1.96𝝈 of the
population mean 
B.
Point estimator = $550
C.
1.96

HI6007 Final Assessment T1 2022


D.
95% confidence interval = Point estimator ± Margin of Error
= 550 ±
(1.96 * 60.05)/sqrt(10)

E.
Lower limit = 550 - (1.96 * 60.05)/sqrt(10) =$512.781
Upper Limit = 550 + (1.96 * 60.05)/sqrt(10) =$587.219

F.
We are 95% confident that the interval contains the population mean.
The interval contains [$512.781 - $587.219] the population mean $550.

G.
i) If the sample size is increased, decreases the width of interval as it decreases the
standard error.

ii) If the sample standard deviation is increased, increases the width of interval as it
increases the standard error.

iii) If the level of confidence is increased, increases the width of interval

Question 4 (11 marks)


The following information has been collected on the sales of greeting cards for the past 6
weeks.
Week Sales ($)
1 85
2 90
3 95
4 110
5 105
6 115

HI6007 Final Assessment T1 2022


A. Develop a linear trend equation that can be used to forecast sales of greeting cards
(6 marks)
t- (t - Sales - (t-Average)*(Sales -
  Week Average Average)^2 Sales Average Average)
  1 -2.5 6.25 85 -15 37.5
  2 -1.5 2.25 90 -10 15
  3 -0.5 0.25 95 -5 2.5
  4 0.5 0.25 110 10 5
  5 1.5 2.25 105 5 7.5
  6 2.5 6.25 115 15 37.5
Sum 21     600   105
Averag
e 3.5     100    

B. Use the linear trend equation developed in part A to forecast sales for week 7. (1
mark)
C. Forecast the sales for week 7 using a three period weighted moving average with
weights of 0.6 (week 6), 0.3 (week 5) and 0.1 (week 4).
(2 marks)
D. Compare and explain why the results in parts B and C are different. (2
marks)

ANSWER:
A.
B1 = 105/21 = 5
B0 = 600/5 – 5(3.5) = 102.5
Sales = 102.5 + 5*Week

B.
Sales for week 7 = 102.5 + 5*7 = 137.5

HI6007 Final Assessment T1 2022


C.
Three period moving average with weights of 0.6 (week 6), 0.3 (week 5) and 0.1 (week
4)
F7 = 0.1*Yweek 4 + 0.3*Yweek 5 + 0.6* Yweek 6 = 0.1*110 + 0.3*105 + 0.6*115
=111.5

D.
Due to the positive trend component in the time series, the trend projection
produced a forecast that is more in line with the trend that exists.

The weighted moving average, even with heavy (.6) weight placed on the current
period, produced a forecast that is lagging behind the changing data.

Question 5 (7 marks)
A. In what situations do we use non-parametric tests and parametric tests? Explain
with at least one example for each. ( 4 marks)
ANSWER a):

Parametric test are used when the test requires any parametric assumption like
normality.
Similarily non-parametric tests are used when do not rely on any parametric
assumption
Due to lack of assumptions, non -parametric test are compared to be less efficient
than parametric test and they may not give accurate solution as they free from
distribution.

Example of Parametric tests:


1. 1 – sample t -test
2. One -Way ANOVA

HI6007 Final Assessment T1 2022


3. 2 sample t -test requires 3 assumptions – Normality, equal variances and
Independence
- Differences in graduate salary based on their gender

Examples of Non – parametric test


1. Mann-Whitney test
2. Kruskal-Wallis, Mood’s median test

B. Compare and contrast the scales on measurements used in statisics. Support your
answer with examples.
(3 marks)
ANSWER b):

The scales of measurement used in statistics are


1. Nominal Scale – The objects are classified based on the list of categories Ex:- data
classified based on gender and nationality
2. Ordinal Scale – A measurement scale which assign value to object based on
ranking with respect to each other. The values are ordered but difference between
values are not important
Ex:- Ranking of the restaurants
3.Interval Scale – The scale represents that one unit indicates the characteristic or
trait on the same magnitude measured across the range of scale. It is an ordered,
constant scale with no zero. For example if BP is measured on a scale , then the
difference between 120 and 180 represent same difference between 60 and 120.
4.Ratio Scale – It is similar to Interval scale but they have true zero points.It is an
ordered, constant scale with natural zero
Ex: - Weight, height , Age

HI6007 Final Assessment T1 2022


Question 6 (7 marks)
For a particular range of cosmetics a filling process is set to fill tubs of face powder with
4 grams on average and standard deviation of 1 gram. A quality inspector takes a random
sample of nine tubs and weighs the powder in each. The average weight of powder is 4.6
grams. What can be said about the filling process, with 95% level of confidence?
Step 1. Statement of the hypotheses (1 mark)
Step 2. Standardised test statistic formula (1 mark)
Step 3. Level of significance (0.5 mark)
Step 4. Decision Rule (Draw a bell to show rejection zone) (2 marks)
Step 5. Calculation of the statistic (1.5 marks)

Step 6. Conclusion (1 mark)

ANSWER:

Step 1
H0 : µ = 4.6
The null hypothesis indicates that the filling process is filling with an average of 4.6 grams
Ha : µ ≠ 4.6
The alternate hypothesis indicates that the filling process is not filling with an average of 4.6 grams

Step 2
Test statistic, Z = (𝒙ഥ-µ0)/(𝝈/ξ 𝒏)

Step 3
Level of significance = 0.05

Step 4

1. Critical value approach


For /2 = .05/2 = .025, z.025 = 1.96
Reject H0 if z < -1.96 or z > 1.96

2. P-value approach

HI6007 Final Assessment T1 2022


Reject H0 if calculated p-value ≤ α = 0.05

Step 5
Test statistic, Z = (𝒙ഥ-µ0)/(𝝈/ξ 𝒏)
𝒙ഥ =4
µ0 = 4.6
𝝈 =1
N =9
Z = (4 – 4.6)/(1/sqrt(9)) = -0.6/(1/3) = -1.8
Test statistic, Z = -1.8
For z = -1.8 , cumulative probability = 0.03593
P – value = 2 * (1 – 0.03593) = 1.92814

Step 6
1.
p – value Approach
Because p-value = 1.92814 > = 0.05, we fail to reject H0 .
There is sufficient statistical evidence to infer that the null hypothesis Is true that
indicates the filling process is filling with an average of 4.6 grams

2. Critical value Approach


Because -1.8 < 1.96, we fail to reject H0.
There is sufficient statistical evidence to infer that the null hypothesis Is true that
indicates the filling process is filling with an average of 4.6 grams

References

HI6007 Final Assessment T1 2022


1. Anon, 2022. What is ANOVA and what can I use it for? Qualtrics AU. Available
at: https://www.qualtrics.com/au/experience-management/research/anova/
2. O'Brien, S. and Yi, Q., 2022. How do I interpret a confidence interval?. Available
at: https://pubmed.ncbi.nlm.nih.gov/27184382/
3. Shorten A, Shorten BHypothesis testing and p values: how to interpret results and
reach the right conclusionsEvidence-Based Nursing 2013;16:36-
37.http://dx.doi.org/10.1136/eb-2013-101255
4.  Isixsigma.com. Available at:
https://www.isixsigma.com/tools-templates/graphical-analysis-charts/making-
sense- time-series-forecasting/

END OF FINAL ASSESSMENT

Submission instructions:
 Save submission with your STUDENT ID NUMBER and UNIT CODE e.g.
NPK1234 HI6007
 Submission must be in MICROSOFT WORD format only
 Upload your submission to the appropriate link on Blackboard
 You have two attempts to submit your assessment with only the final submission
being marked.
Please ensure your submission is the correct document as special
consideration is not given if you make a mistake.
 All submissions are automatically passed through SafeAssign to assess academic
integrity.

HI6007 Final Assessment T1 2022

You might also like