Learning outcomes and pass attainment level:

On completion of the course of study, students should be able to:

1. Demonstrate an understanding of the importance of economic and financial data and the methods of analysis involved in the building and testing of linear regression models and time series models

2. Perform appropriate statistical tests and analysis, interpret the outcomes, and be aware of the strengths and limitations of the approach adopted in the areas of regression and time series analysis.

3. Demonstrate skills in problem analysis

The coursework option consists of data manipulation and estimation in Eviews. There is no lower word limit, your answers need in general to be brief and to the point.

In your answers to the questions below, you should present your Eviews equation estimation output as it would be in published academic papers. (Examine several such papers, the approaches to the presentation are fairly standard.) Raw Eviews regression output should be included only in an Appendix.

You should also include the studies/books you have utilized in your answers in a “References” section. Refer to the attached zipped file titled: “Data for US Market”. The file contains selected data for the US market, including stock markets and economic indicators. There are however prices in a variety of different time segments including weekly, monthly, and yearly data. The pinnacle of financial forecasting is to predict prices such as stock markets or the GDP numbers of a country. Your task is to utilize the data provided and attempt the following.

1. Form a group of up to 4 students.

2. Go through the Eview manual and tutorials to understand how to manipulate data and generate results using the software.

3. Answer the following questions.

Question 1

Go through the zipped file and determine which indicators you can use to predict the Gross Domestic Product of the US by developing an Ordinary Least Square regression equation. Justify why your group decided on this set of indicators and how you would adjust for the different time segments for different indicators. Is your equation created reliable enough for use and how could the accuracy be improved further (you can think of ways beyond the set of data provided)?

Question 2 

In the use of the selected indicators, are issues such as heteroskedasticity, serial correlations, or multicollinearity present? Evaluate how you would go about solving these issues if they are present.

Question 3 

Repeat question 1 but try to predict the S&P500 stock index value instead. Is your equation reliable enough for use and how could the accuracy be improved further (you can think of ways beyond the set of data provided)?

Question 4 

Examine your results in Questions 1 and 3. Would one of the 2 dependent variables be easier to predict compared to the other? Explain why or why not