Project Guidelines
The aim of this project is to test your understanding of and ability to apply the statistical concepts and methodologies discussed throughout the module as well as your ability to analyse and evaluate the outcome of your analysis. The project is deliberately ‘open ended’ or, in other words, not very prescriptive in what and how you should conduct your analysis. You should refer to the material covered in the module and the activities carried out during the term to decide how to answer the questions and shape your investigation. To help your thinking, you can find the following guidelines of some help.
Task 1
In constructing your dataset and in commenting on the data try to think about questions such as: what type of data do you have in the original and in your adjusted dataset? How many variables do you have? What are the types of variables you have? How many countries and observations do you have? Are there variables containing missing observations? How do you handle the missing information? Overall, how would you regard the quality of your data? In investigating the COVID-19 cases and deaths across your continent can you see any pattern or trend?
In addressing the regression analysis make sure to explain how you construct your econometric model by specifying its functional form and its estimated outcome. Make sure to interpret the estimated model, the significance of each individual estimation and the overall goodness of fit of the regression. Produce a clear account of your findings in such a way that WHO officials, who are not necessarily economist and/or statisticians, can understand the meaning of your analysis.
Task 2
The “reproduction rate” is also commonly referred to as the R number. An R number greater than 1 leads to an explosive behaviour in the reproduction of new cases. The dataset contains estimates of the R number for all countries over time. You are asked to carry out an investigation on the ‘likelihood’ that the “reproduction rate” is greater than one. In other words, what factors are likely to influence the probability that the R number will be greater than one? This should be the focus of your analysis: identify those factors that are most likely to make the R number greater than one. Please notice that for this task you are asked to use a cross-section of your database i.e. one observation for each country in your continent at the specified date (25th November 2020).
Task 3
In addressing this question reflect on what type of data and analysis you are asked to carry out. How does it differ from the analysis you carried out in the previous two parts? Make sure to provide a brief but informative analysis of the COVID-19 cases and deaths for the country assigned to you. You should set up your econometrics model and estimate it. Are you, perhaps, considering more than one model because of data availability? As in the previous two parts make sure to comment on your findings both in terms of the estimated coefficient and the goodness of fit. Can you reassure the reader that your estimates are unbiased and efficient? Can you use your model for some forecasting of future COVID-19 cases? Make sure to check that your estimation is providing you with accurate and valid estimates.
