ETF2121 – ETF5912 Data Analysis in Business

ETF2121 – ETF5912 Data Analysis in Business

Assignment 2
Due Date: Friday 8 October 2021 (Week 10)
Due by 11.55pm, Melbourne time
This is an individual assignment worth 15%. Please note that this assignment 2 is due
by Friday 8 October 2021 instead of Wednesday 6 October 2021. You can submit your
assignment early if you wish. Unless otherwise speciÖed, feel free to report any decimal places
you like when presenting your answers. Please include EViews outputs in your answers if
you use EViews.
You are required to complete your answers in a Word document and then upload your
completed Öle in PDF through Moodle submission. Your assignment can be either (i)
entirely handwritten or (ii) entirely typed or (iii) a mixture of handwritten and typed answers.
You may take photos of handwritten answers and paste them on a Word document.
If you prefer not to use Microsoft Word, feel free to try using apps such as Camscanner
(available on iPhones and Android phones) to convert photos of your handwritten answers
directly into PDF. This app can combine multiple pages into one PDF document.
You can access Microsoft Word via the MoVE website https://move.monash.edu/
Please save your Word document as PDF with its Öle name as your name and student ID.
1
Special Consideration
If you are experiencing interferences with your studies that are outside of your control (e.g.
illness, carer responsibilities), you may be eligible for special consideration. You can request
a short extension of up to Öve calendar days by contacting Kew (Chief Examiner) via the
following email: [email protected]
You can request a longer extension of more than Öve calendar days by contacting the university via the following website:
https://www.monash.edu/students/admin/exams/changes/special-consideration
Answer Preparation Tip:
It is very simple to copy output from EViews and paste it directly into a Word document for
writing up your assignment answers. To copy the EViews output, on your computer keyboard
you press PrtScn. Or you can use “snipping tool” to capture screenshots on your Windows
10 – see this website: https://support.microsoft.com/en-au/help/4027213/windows-10-opensnipping-tool-and-take-a-screenshot. Or you can take “screenshot” on your Mac – see this
website: https://support.apple.com/en-au/HT201361.
Question 1 [2.5 marks]
Was there discrimination on the Titanic? Were Örst-class passengers given greater access to
life boats?
The ëunsinkableíluxury liner Titanic collided with an iceberg on her maiden voyage in 1912
and sank with great loss of life. Some passengers on board had paid a very much higher fare
than others for the voyage. A not very subtle sub-text in the most recent Titanic movie was
the notion that the Örst class passengers survived at a higher rate than other passengers.
Imagine we can regard the Örst class passengers as a sample of the total population on board
at the time of the collision. There were 350 Örst class passengers of whom 182 survived the
sinking. In other words our sample proportion p^ = 0:52 while n is 350. The proportion of
survivors for all passengers (total population) on board was p = 0:45:
Using a 0.05 signiÖcance level, test whether the survival rate (proportion of survivors) for
Örst class passengers was greater than that for all passengers. Do the six steps of the test
and interpret the results.
2
Question 2 [2.5 marks]
Each year the personnel department in a large corporation assesses the performance of all of
its employees. Each employee is rated by their supervisors for various aspects of his or her
job on a 4-point scale where 1 = very unsatisfactory and 4 = satisfactory. The manager of
the personnel department believes that the assessment scores last year are lower than this
yearís. To examine the validity of this belief, she randomly draws 50 employees and obtains
both their scores from last year and this year; that is she randomly draws 50 employeesí
scores from last year and the same 50 employeesíscores this year.
The data set is stored in the EViews workÖle performance.wf1. The workÖle includes the
two variables: ìlast_yearîand ìthis_yearî. Some of these data are shown below:
last_year 2 4 1 ……. 1 1
this_year 2 4 3 ……. 1 2
Let the variable ìlast_yearî denote Population A and ìthis_yearî denote Population B.
Do the data allow the manager to conclude at the 5% signiÖcance level that her belief is
correct? You may use EViews to solve this problem.
3
Question 3 [2 marks]
Use the Wilcoxon rank sum test on the data in the following table to determine at the
5% signiÖcance level whether the location of population A is di§erent from the location of
population B.
Sample A 17 20 18 25 16 22
Sample B 17 25 33 38 15 26 21
Question 4 [3 marks]
The Örst regression was carried out in 1885 by Sir Francis Galton who wanted to see the
relationship between parentsí height and the height of their children. Sir Galton thinks
that motherís height should be a good determinant of childís height. The EViews workÖle
height.wf1 contains data gathered on the heights of 103 children and their parents. The
workÖle includes the three variables: child = childís height, mother = motherís height and
f ather = fatherís height, where height is measured in centimetres.
(a) Using EViews, plot the data of child (y-axis) and mother (x-axis) in a scatter diagram.
Is there a relationship? Explain. [1 mark]
(b) Using EViews, calculate (i) the coe¢ cient of correlation between child and mother,
(ii) the coe¢ cient of correlation between child and f ather, and (iii) the coe¢ cient of
correlation between mother and f ather: What do these statistics tell you? [2 marks]
4
Question 5 [5 marks]
(This question considers a possible implication of the previous height question). One general
belief held by observers of the business world is that taller men earn more money than shorter
men. In a study reported in the Wall Street Journal published on 18 March 20121
, 29 MBA
graduates, all about 30 years old, were surveyed and asked to report their annual incomes
(in thousands of dollars) and their heights (in centimetres). This information is recorded in
the EViews workÖle earn_more.wf1.
(a) Use EViews to perform a regression analysis where the dependent variable is the annual
incomes. Write down the sample regression line. Express your answers in 4 decimal
places (e.g. round 1:23456 to 4 decimal places is 1:2346 and e.g. round 1:23451 to 4
decimal places is 1:2345) [1 mark]
(b) Interpret the estimated intercept and slope coe¢ cients. [2 marks]
(c) Do you think that this regression model gives a reliable prediction of income for a man
175 centimetres tall? If not, explain why not. If so, explain why and predict the income
of a man 175 centimetres tall with an MBA. [1.5 mark]
(d) What percentage of the variation in incomes is explained by heights? [0.5 mark]
1For those who are interested, the link is https://www.wsj.com/articles/
SB10001424052702304459804577285710102760818. You need to subscribe to continue reading. However
reading this article will NOT at all help you to answer the questions.