Scribd
Upload
20 views3 pages

QBUS5002 Chapter 9 Tutorial (Question) TW

Uploaded by

zhewang1903
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views3 pages

QBUS5002 Chapter 9 Tutorial (Question) TW

Uploaded by

zhewang1903
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

University of Sydney Business School

Discipline of Business Analytics


Tutorial 9

Chapter 9: Statistical Inference IV & V

Date file: QBUS5002 Chapter 9 Tutorial (Data).xlsx


Solution file: QBUS5002 Chapter 9 Tutorial (Solution).xlsx

Online Activities
• Revision on main concepts – ANOVA, HT of normality and HT for the equality of two
population variances
• Tutorial questions
• Complete two tasks
• Q&A

1. Tutorial Questions

Use Excel and QBUS5002 Analysis ToolPaks to perform hypothesis tests in the following questions.

1. [Tutorial 8, Q1] Weekly percentage returns of the CBA and ANZ stocks are collected from 1 Oct
2017 to 23 Sept 2018. Suppose that the (population) means and standard deviation are µ1 and σ1
(for CBA) and µ2 and σ2 (for ANZ).
(i) Test, at the 5% level of significance, whether the weekly percentage returns of the two stocks
are normally distribution. Do you have the same conclusion at the 10% level of significance?
(ii) Test, at the 5% level of significance, whether the weekly percentage returns of the two stocks
have the same variance.
(iii) Test, at the 5% level of significance, whether the weekly percentage return of CBA is smaller
than that of the ANZ.

2. [Tutorial 8, Q2] A financial institution requires its new employees to attend a new training course
three months after they join the company. This is a training course to improve employee’s
understanding of the company’s mission, values, directions, operations, and clients, etc. Before
attending this training course, the Human Resources Manager asks each new employee to complete
a questionnaire and a score, X, is obtained. After attending the training course, the employee will
complete a different questionnaire and another score, Y, is obtained. The Human Resources
Manager takes a random sample of 50 new employees and their scores are obtained. Test, at the 5%
level of significance, whether the difference of the scores follows a normal distribution.

QBUS5002 Quantitative Methods for Accounting p.1


University of Sydney Business School
Discipline of Business Analytics
Tutorial 9

3. [ANOVA & Test for Normality] The Personal Investment Money Guide regularly publishes the top
300 Australian companies ranked in descending order of market capitalisation. The data, given in
Spreadsheet Q3”, are the yields of 50 companies selected at random from the three groups of 100
companies (Note: We will not apply finite population adjustment here). An investment consultant
would like to know whether the average yields of the three groups are different.
(i) What conclusion can you draw from the data regarding the average yields of the three groups
companies? Use  = 0.05.
(ii) State the assumptions that you have made and test, at a 5% level of significance, whether
these assumptions are valid.
(iii) If not all assumptions in (ii) are valid, does the result in (i) valid? Comment.

Note: This example demonstrates the impact of outlier(s) on statistical inference.


The spreadsheet also contains data for net income growth (average % growth in net value) and book
value growth (average % growth in book value) over the 5-year period and you can perform similar
analyses on these data.

2. Complete two Tasks

Task 1: ANOVA + Check for model assumptions (2 test for normality + F test for equality of two
populations variances)

An auditor takes a random sample of size 25 from each of three normal distributions, N(i, i2), i = 1,2,3
(NB: this is an assumption) and these random samples are independent. He wants to know whether the
means of the three populations are different. Use a 5% level of significance in all statistical tests.
(i) Perform a one-way ANOVA to test whether the population means are equal. If the null hypothesis is
rejected, which means are likely to be equal and means which are unlikely to be equal (without
performing additional statistical analysis such as multiple comparison).
(ii) The one-way ANOVA model assumes that the random errors (and hence the data) are normally
distributed. Test whether this assumption is valid or not.
(iii) The one-way ANOVA models assume that the random errors have a constant variance. Test
whether this assumption is valid or not.

Task 2: To verify the Central Limit Theorem using simulation


The CLT states that the sampling distribution of the sample mean of a random sample of size n, n ⩾ 30
from any probability distribution (except normal) with mean  and variance 2 can be approximated by
the normal distribution with mean  and variance 2/n. Let’s verify whether or not the sample mean is
normally distributed using simulation.

QBUS5002 Quantitative Methods for Accounting p.2


University of Sydney Business School
Discipline of Business Analytics
Tutorial 9

Consider the following discrete probability distribution.

x 0 1 2 3

P(X=x) 0.1 0.4 0.3 0.2

A random sample of size 50 is taken from this distribution and the sample mean is calculated. Repeat this
experiment for another 399 times so that you have a random sample of 400 sample means for normality
testing.
Instructions: Generate 400 sets of 50 values from this probability distribution in Excel using “5002” as
random seed if required. Note: Values simulated from Excel using PC and Mac computers are different.
Test, at the 5% level of significance, whether the sample mean is normally distributed. Also check
whether the mean and variance of the sample mean in this random sample are close to their theoretical
values.

QBUS5002 Quantitative Methods for Accounting p.3

You might also like