Homework answers / question archive / ECO4000 Final Project The goal of the (individual) project is to have students apply the tools of regression analysis to a problem of interest

ECO4000 Final Project The goal of the (individual) project is to have students apply the tools of regression analysis to a problem of interest

Economics

Share With

---

ECO4000 Final Project

The goal of the (individual) project is to have students apply the tools of regression analysis to a problem of interest. It could be an economic or financial issue that you would like to understand in more detail or any other context where the LRM can be applied.

1. The first step is to identify a question of interest, in particular a Y (dependent) variable that you would like to describe and some (independent) variables X that you use to explain Y .

It is important that you are able to discuss what economic theory (or common sense) would suggest the relationship between Y and the Xs should be.

This will give you a framework to interpret the regression results. Remember that:

• cross-sectional data consists of different units observed at the same point in time (eg, a sample of workers in 2019); in this case your goal is to explain why Y is different across units

• time series data is when we track one variable over time (eg, GDP from 1950 to 2019 at the quarterly frequency); here the goal is to explain the fluctuations over time of the variable

2. The following step is to find a suitable dataset that provides a sample for the variables you want to observe. There are lots of open datasets that can be freely downloaded on the internet. Make sure that you quickly identify a dataset, otherwise it is advisable that you change to a different project idea for which you can find data.

3. Once you have the idea and the data, it is time to apply the LRM. Some issues you should investigate and discuss are: • Are the variables that you thought could be important statistically significant (at 5%)? • Are there variables that you would have liked to include in the regression but could not find data for?

Discuss whether you expect these omitted variables to introduce bias in your OLS estimates. • Are the estimated effects (betas) of the expected sign?

• Interpret the estimated effects

• Are there one or more variables that you would like to exclude from the regression because they are irrelevant? Use the F-test • Include at least one type of nonlinearity in the LRM and interpret its effect 4. Prepare a presentation with the following slides: • Slide 1: title of the project, author • Slide 2: Describe the question you are trying to answer and why it is relevant

• Slide 3: Describe the empirical LRM in terms of the dependent variable, independent variables, and formulate an expectation of the sign of their relationship; include at least one nonlinearity in the LRM that you are considering

• Slide 4: Discuss data sources

• Slide 5: Discuss the estimation results in terms of the finding on the relation between the Xs and Y (and their significance), the goodness-of-fit of the regression, the relevance of the nonlinearity included in the model, and the questions discussed earlier

• Slide 6: draw some conclusions from your analysis, in particular discussing the answer to your initial question 1 5. In addition to the presentation, prepare a 5 minute video in which you discuss your analysis using the slides (this can be done by starting a Zoom meeting and recording your presentation of the slides or by voiceover narration in Powerpoint). If you use Zoom I recommend that you record to the cloud or store the presentation in a cloud service so that you only need to share the link in Blackboard.

6. Upload in Blackboard (Exams/Final Project) the following: 1) presentation, 2) link to the video, 3) excel spreadsheet with data and regression results.

7. Deadline to submit the final project is May 21th, 2021 by 11.59pm.

8. Weight of the final project is 10% of the overall grade 1 Assessment The assessment is based on the following criteria: • The research question is relevant and well-specified

• The dataset is suitable to answer the research question

• The econometric analysis is competently executed • The conclusions are well-grounded on empirical evidence provided

• Clarity of the video presentation 2 Academic Honesty In addition to the standard rules of academic honesty, you should be aware that I’ll scan every project with the plagiarism detector software (and in case of significant matching I reserve the right to give a 0 out of 100 on the final project).