# FE7066 Data Analysis for Global Business

Assignment Component 002 Coursework

**Weighting 70%**

**Analysing quantitative secondary data**

This assignment component comprises a report of 2,000 words. A combination of tasks such as investigating relationship of variables, testing of hypotheses and forecasting will be included in the report.

**Deadline**

5^{th} May 2023 by 3:00 p.m.

**Submission**

Students to submit their work via Online Submission of module Weblearn.

**Requirements**

There are two questions with detailed requirements. Question 1 carries 60% and question 2 carries 40%. Both two questions are compulsory, and students should provide answers to each question accordingly in their report.

**Question 1 Investigating relationship of variables and testing of hypotheses (1,300 words)**

In the spirit of Arbitrage Pricing Theory (APT), you are required to examine regressions that seek to determine whether the monthly returns on Microsoft stock can be explained by reference to unexpected changes in a set of macroeconomic and financial variables.

You are provided with an EViews work file: macroassessment002. The sample period is from March 1986 to March 2018.

You are required to complete the following tasks and report the results:

Run the following OLS regression and present your output table:

**ermsoft c ersandp dprod dcredit dinflation dmoney dspread rterm**

Using the 10% significance level, which of the variables has coefficients that are statistically different from zero? (3 MARKS)

Explain the null hypothesis of the F-test of the regression and state whether the null is rejected. (3 MARKS)

You are given the following null hypotheses: the coefficients of DPROD, DCREDIT, DMONEY, and DSPREAD are zero. Test the null hypotheses that the parameters on these four variables are jointly zero using an F-test.

How many restrictions are there? (2 MARKS)

How do you express the restrictions in EViews? (3 MARKS)

Present your output table and explain whether the null hypotheses are rejected. (3 MARKS)

R-squared and adjusted R-squared

Explain what R-squared measures.** **(3 MARKS)

Explain the difference between R-squared and adjusted R-squared

(3 MARKS)

Delete DPROD from the original OLS regression and present the output table. Comparing with the original regression, does the R-squared increase? Does the adjusted R-squared increase? Based on the comparison, should DPROD be included in the regression?

(3 MARKS)

You want to test whether the residual from the regression is homoscedastic using the White’s test. Note that we exclude the White cross terms

What is the null hypothesis?** **(2 MARKS)

Present your output table and explain whether the null is rejected.

(3 MARKS)

You then want to check whether the residuals are autocorrelated by looking at the Durbin-Watson test:

What is the Lower and upper critical values? (3 MARKS)

What is the null hypothesis?** **(3 MARKS)

Does the Durbin-Watson test statistics suggest the null is rejected?

You then want to employ the Heteroscedasticity and Autocorrelation Consistent (HAC) estimator.

Present your output table and explain whether your conclusion in step a) is still valid. (3 MARKS)

You want to see whether multicollinearity is an issue amongst regressors by looking at the correlation matrix. The regressors in the original equations are **ersandp dprod dcredit dinflation dmoney dspread and rterm**.

Present the correlation matrix. (2 MARKS)

Are there any regressors that are highly correlated with others (using 0.80 as a criterion). (3 MARKS)

You now run the normality test of the residuals using the histogram-normality test.

Present your results.** **(2 MARKS)

From the histogram, is the residuals normally distributed?** **

(2 MARKS)

What command do you use to generate these two dummy variables? (3 MARKS)

You then include these two dummy variables into the following original regression**:**

**ermsoft c ersandp dprod dcredit dinflation dmoney dspread rterm**

to produce an output from this Multiple Regression with Dummy Variables.

and** **you carry out again the normality test of the residuals

Present the output of the histogram-normality test. (2 MARKS)** **

From the histogram, is the residuals normally distributed?

(2 MARKS)

You now go back your original regression as in step a).

You want to examine the parameter stability using the Chow test. You choose a chow breakpoint of 1996 month 3 (March) by entering **1996m01** or **1996:01 **in the dialogue window.

Present your output table. (2 MARKS)

Based on the Chow test, does the F-statistic suggest that the null is rejected? (2 MARKS)

**(60 MARKS IN TOTAL)**

**Question 2 Test of cointegration and Error Correction Model (ECM)**

**(700 words)**

You are required to collect the weekly closed market share price of the SSE Plc (SSE) and the FTSE 100 Stock Index (UKXX), for the period from Friday 1^{st} March 2013 to Friday 3^{rd} March 2023.

Generate the logarithms of the share price of the SSE Plc (LSSE), and the FTSE 100 stock price index (LUKXX), for the given period.

(2 MARKS)

Produce a Graph of the series, LSSE and LUKXX. (3 MARKS)

Assume both series are I(1) and carry out an Engle-Granger cointegration test. (15 MARKS)

Using Error Correction Models (ECMs), model the relationship between the series. (20 MARKS)

**(40 MARKS IN TOTAL)**

**Assessment Criteria**

The assignment will be assessed on the following:

Demonstration of an understanding of the analysis of secondary quantitative data.

The ability to apply computational skills and use regression to analyse relationship between variables.

Quality of interpretation of the outputs.

Demonstration of critical and analytical thinking in discussion and the drawing of conclusions.

Clarity of the report and quality presentation of the data used.

**Presentation of your coursework**

You are expected to include answers to the questions with EViews outputs as required.

You should submit your coursework with an official cover page which you can generate and download from your Evision. You may also produce your own cover page which may include information such as the assignment component number of this coursework, module code, module title, name of the school and date for submission. You should **NOT** display your name on this cover page and any other pages of your coursework.

You should choose Times New Roman with font of size 12 and double spacing to present your work.

**Warnings**

The coursework submitted should be your own work.

The use of any other person’s work constitutes plagiarism and is an assessment offence.