This group assignment requires your team to apply statistical knowledge and skills learned from BEO5539 lectures between week 2 and 7. Please note the following: (a) Each assignment group will comprise no more than 3 members. Each team member is expected to work collaboratively with one another and contribute their fair share of effort in completing this group assignment. Team members must be from the same tutorial group. It is important that there is coordination among team members to ensure this assignment is duly completed and submitted by the deadline.

(b) Real data must be used for this assignment. The variables for this group assignment are identified in Table 1 below. These variables can be extracted from Orbis database via http://library.vu.edu.au/search/f?SEARCH=orbis&sortdropdown=-

(c) In Orbis database, please:

– select 3 countries (at least one developed country and one developing country);

– select one industry (same industry for the 3 countries);

avoid selecting data from 2014 because significant amount of the 2014 data is yet to be released through this database. Missing data must not be included in the analysis and therefore must be deleted from the spreadsheet. – note that the listed companies extracted from Orbis database are used to infer about the population of companies in this assignment.

(d) Your sample sizes will vary according to your choice of countries and industry for focus. If any of the sample size (for a country) is less than 30, please discuss this with your instructor. (e) Some of the variables may have to be transformed or re-coded before the application of a statistical technique.

(f) Please use SPSS for statistical analysis in this group assignment.

Relevant SPSS statistical output must be properly analysed and interpreted. Countries comparison for the same industry is usually required (unless explicitly stated) in the questions. In the tests of hypotheses, ensure that the p-values are identified in your statistical decision. (g) Please provide a number for every table, graph or figure used and make clear reference to the table/graph/figure in your discussion.

(h) The word limit for this group assignment is 5,000 words (excluding SPSS output). (i) This assignment is due in week 8. Please submit your assignment in Word document via VU Collaborate together with a completed assessment declaration form signed by all team members and your SPSS data set.

(j) It is not compulsory to use database beyond Orbis database. However, students may want to harvest data from other databases and used them in conjunction with those harvested from Orbis in the form of a merged data file, or use an entirely different database. In these situations, students should approach the instructor for approval before using their preferred database(s) for their group assignment. The instructor will discuss with the team to find out whether the variables and data format are suitable for statistical analyses in the group assignment.

Table 1 – Variables from Orbis database for Assignment 1

Variables

V1. Country ISO Code

Remarks

This variable will automatically appear in the

spreadsheet when you select your 3 countries.

Each country will have a different label.

V2. Operating Revenue

Definition: Income derived from sources related

(Turnover)

to a company’s everyday business operations.

V3. BvD Indep. Indic.

Known as ‘company independence indicator’. It

categorizes the degree of independence of a

company. The indicator marks are as follows:

A: No recorded shareholder with more than 25%

direct or total ownership.

B: No recorded shareholder with more than 50%

direct or total ownership. And one or more

shareholders are recorded with more than 25%

direct or total ownership.

C: One recorded shareholder with more than

50% direct or total. Also given to a company

when a source indicates that the company has

an ultimate owner.

D: One recorded shareholder with a direct

ownership of over 50%.

U: Unknown status of independence.

V4. Returns on earning (using

Definition: Profit & loss before tax divided by

Profit & Loss before tax) in %

shareholders’ funds. It measures how much

profit a company generates (or loss made) with

the money shareholders have invested.

V5. Solvency ratio in %

Definition: Shareholders’ funds divided by total

assets. It measures whether a company can stay

solvent.

V6. Number of companies in a In this assignment, use this variable as a proxy corporate group.

for the size of a corporation.

You may wish to select a set of Please ensure your selected variables make different variables from other sense in context, otherwise your assignment may end up dealing with spurious relationships

databases. Please discuss

among your variables.

your preferred variables with

your instructor.

Assignment tasks:

The variables for this assignment are as follows:

V1) Country ISO Code. Recode this variable into the following: 1 = Name of country 1

2 = Name of country 2

3 = Name of country 3

Note: Attach value labels to this variable.

V2)Operating Revenue (Turnover) in dollars value.

V3) BvD Indep. Indic.

This variable categorizes the degree of independence of a company. The indicator marks are as follows:

A: No recorded shareholder with more than 25% direct or total ownership. B: No recorded shareholder with more than 50% direct or total ownership. And one or more shareholders are recorded with more than 25% direct or total ownership. C: One recorded shareholder with more than 50% direct or total. Also given to a company when a source indicates that the company has an ultimate owner. D: One recorded shareholder with a direct ownership of over 50%. U: Unknown status of independence.

Note: This variable must be re-coded into a 4 categorical data (i.e., C and D have to be re-coded into the same category).

V4) Returns on earning (using Profit & Loss before tax) in % V5) Solvency ratio in %

V6) Number of companies in a corporate group. The value ‘0’ indicates the company is a standalone company – without branch, subsidiary or affiliated company. 1) Week 2 topic – Descriptive Statistics

Present V2 to V6 for your selected industry using the following descriptive statistics* and compare them across countries:

Topics

Week 2: 1A) Descriptive Statistics – presenting data in tables & charts Frequency, relative frequency, and cumulative frequency.

Histogram/column chart/pie chart (select 2 measures)

Week 3: 1B) Descriptive Statistics – numerical descriptive measures Central tendency (mean, median)

Non-central tendency (80 percentile, 1st quartile)

Variability (range, variance, standard deviation, coefficient of variation and interquartile range) *Note: Some of the data has to be transformed or re-coded before certain descriptive statistics could be used. On the other hand, there are data that cannot be computed for some descriptive statistics.

2) Week 3 topic – Estimation

2a) Construct a confidence interval to estimate the population mean at 99% confidence interval for this industry in each country for V2, V4, V5 and V6. Interpret each confidence interval in context and make a comparison across countries in your analysis.

2b) Construct a confidence interval to estimate the population proportion of companies with ‘No recorded shareholder with more than 50% direct or total ownership, and one or more shareholders are recorded with more than 25% direct or total ownership’ (refer to V3) at 95% confidence interval for this industry in each country.

Recode V3 into different variable with:

B=1

Others =0

Interpret this confidence interval in context and make a comparison across countries in your analysis.

3) Week 4 – Hypothesis Tests

Requirements for each hypothesis test:

i)

Formulate the null and alternative hypotheses.

ii)

State your statistical decision using significant value (?) of 5% for each test. iii)

State your conclusion in context.

3a) Perform a one-sample test:

Variable

Hypothesis statement for test:

V4: ROE – returns Test whether the population’s average ROE for this industry is different from the on earning (using combined average ROE of the 3 countries.

Hint: Enter the total average ROE of the 3 countries into the test value field in profit & loss before tax)

SPSS.

3b) Perform independent samples test:

Variable

Hypothesis statement for test:

V2: Operating

Test whether the population mean of operating revenue (turnover) for this Revenue

industry is different for each pair of countries.

(Turnover)

V3: BvD Indep.

Focusing on only companies with known status of independence, test whether Indic.

the population proportion of companies with no recorded shareholder with more than 25% direct or total ownership (A) is the same for each pair of countries.

Hint: Do not include companies with unknown status of independence in this test. Re-code this variable in the filtered sample into:

A=1

Others = 0

V5: Solvency

Test whether the population mean of solvency ratio for this industry is the same ratio for each pair of countries.

3c) Perform paired samples test:

This test requires you to focus on all companies in this industry located in a country (select one of the three countries) in two time periods. Identify a major event to determine whether this event makes a significant impact on V2 and V4. For example, you wish to determine whether the Global Financial Crisis (GFC) in 2007 has made a significant impact on V2 and V4, you will need data from 2006 and 2008 for all companies in this country for V2 and V4. Hint: Please make sure there is no missing data for V2 and V4 in 2006 and 2008 for all companies in your sample.

Variable

Hypothesis statement for test (based on GFC example):

V2: Operating

Test whether the population mean of operating revenue (turnover) in 2008 is Revenue less than 2006 for this industry in your selected country.

(Turnover)

V4: ROE – returns Test whether the population mean of ROE in 2008 is less than 2006 for this on earning (using industry in your selected country.

profit & loss before tax)

4) Week 5 – Non-parametric tests

The following tests are non-parametric equivalent of independent samples and paired samples tests done earlier in 3(b) and 3(c).

Requirements for each hypothesis test:

i) Formulate the null and alternative hypotheses.

ii) State your statistical decision using significant value (?) of 5% for each test. iii)

State your conclusion in context.

iv) Compare each non-parametric test with its equivalent parametric test, and determine which is an appropriate test.

4a) Conduct Wilcoxon rank sum test for these variables:

Variable

Hypothesis statement for test:

V2: Operating

Test whether the population distribution of operating revenue (turnover) for Revenue this industry is different for each pair of countries.

(Turnover)

Compare this test with independent samples test and determine which is an appropriate test (explain why).

V3: BvD Indep.

Focusing on only companies with known status of independence, test whether Indic.

the population proportion of companies with no recorded shareholder with more than 25% direct or total ownership (A) is the same for each pair of countries.

Hint: Do not include those companies with unknown status of independence in this test. Re-code into:

A=1

Others = 0

Compare this test with independent samples test and determine which is an appropriate test (explain why).

V5: Solvency

Test whether the population distribution of solvency ratio for this industry is the ratio same for each pair of countries.

Compare this test with independent samples test and determine which is an appropriate test (explain why).

4b) Conduct Wilcoxon signed rank test for these variables:

Variable

Hypothesis statement for test:

V2: Operating

Test whether the population distribution of operating revenue (turnover) in Revenue 2008 is less than 2006 for this industry in the same country you selected in 3(c). (Turnover) Compare this test with paired samples test and determine which is an appropriate test (explain why). V4: ROE – returns Test whether the population distribution of ROE in 2008 is less than 2006 for on earning (using this industry in in the same country you selected in 3(c). Compare this test with paired samples test and determine which is an profit & loss appropriate test (explain why).

Requirements:

(1) Conduct a one-way ANOVA to test the null hypothesis that means operating revenue(V2) are the same for all three countries (V1) amongst the population of companies.

(2) Find Tukey confidence intervals for the mean difference between pair of countries. Is there any statistically significant mean difference between pair of countries? (Report on these only where the differences are statistically significant.)

(1) Conduct a one-way ANOVA to test the null hypothesis that means operating revenue(V2) are the same for all four company independence indicators(V3) amongst the population of companies.

(2) Find Tukey confidence intervals for the mean difference between pair of indicators. Is there any statistically significant mean difference between pair of indicators? (Report on these only where the differences are statistically significant.)

(1) Conduct a 2-way ANOVA to test the null hypothesis that means operating revenue(V2) are the same in all three countries (V1) and four company independence indicators(V3) amongst the population of companies in this industry. Allow for the possibility of an interaction effect between country and indicators towards operating revenue.

(2) Produce Tukey confidence intervals for row and column differences. (Report on these only where the differences are statistically significant.)

6) Week 7 – Randomised block design, Kruskal Wallis test, & Correlation Requirements for Randomised Block Design and Kruskal Wallis tests: iv)

Formulate the null and alternative hypotheses.

State your statistical decision using significant value (?) of 5% for each test. vi)

State your conclusion in context.

5(b) Randomised Block Design:

Variable

Requirements:

Use the randomized block design data to test the null hypotheses ● V1: Country ISO that the means operating revenue are the same for each of the Code treatments (V3) and the same for each of the blocks (V1).

● V2: Operating

Compare this test with one-way ANOVA test in 5(a)(i) and

Revenue (Turnover)

● V3: BvD Indep. Indic. determine whether including the block effect increases the power of the test.

5(b) Kruskal-Wallis:

Variable

● V1: Country ISO

Code

● V2: Operating

Revenue (Turnover)

5(c) Correlation:

Variable

● V3: BvD Indep. Indic.

● V6: Number of companies in a corporate group

Requirements:

Use a Kruskal-Wallis test of the null hypothesis that the distributions of operating revenue are the same for all three countries. Compare this test with one-way ANOVA test and

determine which is an appropriate test (explain why).

Requirements:

Conduct a correlation test for these two variables using either Pearson or Spearman correlation. Explain the rationale behind your choice of correlation technique.