STATISTICS AND ECONOMETRICS
2009/10 LEVEL 2
Assessment: Project Work
代写留学生论文Submission Date: Wednesday 5 May 2010, Until 3 p.m.
Undergraduate office, Pemberton Building
Failure to submit TWO copies of your work by this deadline will result in a mark of 0%
The assignment must also be submitted in electronic format to the ‘Turnitin’ drop-box in the Blackboard site for this module and the Turnitin reference code added to the hand-in sheet
You are advised to plan your work carefully and back-up your work. Computing and printing problems will NOT be accepted as reasons for non-submission
For your information when preparing your coursework essay, refer to the following:
o University Regulation on the Presentation of Work for Formal Assessment (URL: www.brad.ac.uk/admin/acsec/QA_Hbk/QAH5_19.html)
o University Statement on Academic Integrity (URL:www.brad.ac.uk/admin/acsec/assu/statement_on_academic_integrity.htm)
o Student Handbook on Plagiarism - remember to fully reference your work in the Harvard format.
This study requires each student to compile a sample of observations for 15 countries:
for the period of 2007 or 2008.
Each student will be given a specific set of countries which they must use.
DATA:
(1) QMSE= quality of maths and science education
Data for this on a 1-7 point scale can be obtained from the world economic forum at: http://www.weforum.org/en/initiatives/gcp/Global%20Competitiveness%20Report/index.htm where you can download the Global Competitiveness Report (GCR).
(2) Real GDP per capita
You can use the country profile figure in the complete Global Competitiveness Report which is for 2008 although other sources are also available.
QUESTIONS:
1. Present the data in a table showing the names of the variables. Make sure the full definitions and sources of each variable are given.
[15 marks]
2. The first equation to be estimated is:
ln(GDP i) = b0 + b1 QMSEi + ui
where the variables correspond to those listed above. ln stands for natural logarithm
and the subscript i represents the observation (country).
Answer these questions:
(i) In terms of economics, how would you describe the hypothesis that this equation will test?
(ii) What sign do you expect b1 to be and why?
(iii) Explain precisely the meaning of a one unit increase in QMSE in this model.
(iv) ‘b0 has to be positive because QMSE is positive’. True or False? Explain.
(v) Why is the final term (u) included in the equation?
[10 marks]
3. Estimate the above equation by OLS and present the results in a suitable table
(N.B. marks will be lost for simply pasting over the computer output)
(i) Comment on the plausibility of your b1 value with respect to your answer to 2(ii).
(ii) Comment on the R squared statistic for this equation.
(iii) Using your estimated equation work out how much of a change in real GDP would be predicted to occur if a country went from a score of 3 to a score of 5 on QMSE.
(iv) Explain what assumptions need to be made about the u term in order to carry out t-tests on the parameters.
[20 marks]
4. You are now going to use the following equation
GDPi = b0 + b1 QMSEi +b2 QMSEi 2 + ui
(i) Estimate this equation by OLS and present the results in a suitable table.
(ii) Interpret the results specifically indicating whether the relationship appears to be u-shaped, inverted u-shaped or of some other pattern.
(iii) /Does your equation have a turning point? Explain.
(iv) Is the R2 from this equation higher than that from the previous equation? Is this information of any use?
(v) Repeat the question in 3(iii) for this functional form and compare your answer with the answer obtained there.
[15 marks]
5. Carry out the following hypothesis tests for THE SECOND EQUATION ONLY.
(i) H0 : b0=0 against the two sided alternative at the 10% level.
(ii) H0 :b1=0 against the two sided alternative at the 5% level.
(iii) H0 : b2 <0 against the alternative at the 10% level.
(iv) H0 : b1= b2=0 against the alternative at the 5% level.
[10 marks]
6. Examine the residuals of the second equation and indicate whether any of your nations are outliers in terms of the predictive accuracy of the regression equation.
[10 marks]
7. You should now write a short report of 450-600 words. This should briefly summarize your findings, but most of this answer should consist of further exploration of your data and suggestions for improvement of the model you have estimated. The full GCR report is 513 pages long and provides many sources of data from which you could obtain new variables.
[20 marks]