#### Econometrics

**Paper , Order, or Assignment Requirements**

Learning outcomes to be examined in this assessment

- a) Design a simple econometric research project containing a full range of techniques
- b) Evaluate basic techniques of econometrics in relation to specified problems
- c) Discuss the role and limitations of econometric methods in the analysis of contemporary problems.

Percentage of marks awarded for module:

The assignment is worth 100% of the total marks for the module.

Assessment criteria

Explanatory comments on the assessment criteria

Maximum marks for each section

Content, style, relevance,

Content will reflect the students’ ability to understand and to analyse econometrics as taught.Answers to

50%

Page 2 of 6

originality be based on an interpretation of class hand-outs and evidence of background reading. That is, there should be clear demonstration of focused rigorous research from recognised authoritative sources.

Format, referencing, bibliography

- The limit is not a guide, it is an instruction. • Do not stray into tangential material. Always ask yourself – is what I’ve written relevant to the question set? • Copy and Paste. Please try to refrain from doing this. • Font size. Please use font size of approx.. TNR 12, Arial 12 size. • Double space. • Read (aloud) what you have written. If we can’t understand your argument, you will not get any credit for it.

The Study Skills Handbook and Module provide detailed guidance on referencing. Ensure you reference your citations using the Harvard method.

10 %

Constructive critical analysis, introduction, conclusion

Demonstration of a clear understanding of the issues. Use of academic models. Clear focussed understanding of a topic.

Critical analysis is an important test of the student’s ability to evaluate econometric concepts.

Introductions and conclusions should briefly address the issues to be discussed and discussed

40%

Page 3 of 6

respectively.

- Candidates must clearly label their ID Number on additional separate reference, formula or answer sheets.
- We will support Gretl for use in econometric estimation in this course, but you are welcome to use other software for estimation. However, please note Gretl is the software which is installed in the computer labs.Gretl is an open source and free software that can be downloaded and installed in PC, Mac and Linux machines.
- Please note that academic misconduct, such as collusion and ghost-writing, will NOT be tolerated. The module leader and tutors reserve the right to call the students to an informal hearing in order to query about specific details of the assignments.

Page 4 of 6

Assignment

Question 1 (40 marks)

A socio-economist wants to know if countries that are more democratic are also having more trusting societies. Considering ‘Trust’, the average trust level in a country (expressed as a percentage of people stating that they trust their neighbours) and ‘DemocracyIndex’, a measure for how democratic a country is (using the Economist Intelligence Unit Index in 2015), the following OLS estimation has been undertaken by the researcher.

OLS, using observations 1-63 Dependent variable: Trust

coefficient std. error t-ratio —————————————————– const 22.6191 0.0709278 3.189 DemocracyIndex 1.112178 0.00985170 1.139

- Find information on the Economist Intelligence Unity Index for 2015 (www.eiu.com) and describe the data. What are the constituents of this index and find out the democracy score of the UK and Island in 2015. [5 marks]
- Write down the regression results in equation form. [5 marks]
- Interpret the results. Briefly explain if they make sense? [10 marks]
- What is the expected trust level for the UK? How would the expected trust levels change for a countries with more authoritarian governments than the UK? [5 marks]
- Establish the 95% confidence interval for the estimated coefficient of DemocracyIndex. [5 marks]
- What is the p-value that is associated with the estimated coefficient of DemocracyIndex? [5 marks]
- Interpret the R-squared statistics of this model. [5 marks]

[Max: 40 marks, Learning Outcome: b]

Page 5 of 6

Question 2 (40 marks)

The researcher has ‘amended’ her regression by adding the variable “Gini”, which measures the income inequality of a country by utilizing the Gini Coefficient.

OLS, using observations 1-63

Dependent variable: Trust

coefficient std. error t-ratio

—————————————————–

- Find information on the Gini Coefficient (data.worldbank.org/) and briefly describe how the coefficient measures income inequality in a country. Explain if you think this is an appropriate measure for income inequality. [5 marks]
- Test for significance for both Gini and DemocracyIndex at the 10% level and interpret the results. Explain if countries that have high levels of income inequality are less or more trusting. [5 marks]
- Estimate UK’s trust level by using UK’s actual Gini Coefficient of 32.5% and its Democracy Index of 8.31.Note that the UK’s actual trust levels are 30.0% (Source: World Value Surveys). Compare the actual figure with your estimate and explain why the estimate differs from the actual figure. [10 marks]
- Compare the results with the regression is Question 1. Discuss why the estimated coefficient of DemocracyIndex has decreased in the amended regression. [10 marks]
- Has the significance of DemocracyIndex increased or decreased after the inclusion of Gini? Explain the results. [5 marks]
- What other two variables can you think of that could be included in the model to predict trust levels. Explain why the variables would be a good choice and gauge their expected sign. [5 marks]

[Max: 40 marks, Learning Outcomes: b, c]

Page 6 of 6

Question 3 (20 marks)

“In many cases, two time series processes appear to be correlated only because they are both trending over time for reasons related to other unobserved factors” Jeffrey M. Wooldrige (2014: 294)

Discuss Wooldridge’s statement with regards to the limitations of OLS when using time series data. How can a researcher reduce the danger of producing false causal inferences here?Support your arguments with at least one example from the peer reviewed literature. (Note: The wordlimit for this question is 800 words + or – 10%)

[Max: 20 marks, Learning Outcome: c]

Total marks for Report: 100