Choose a topic and hypothesis of your choice, find an appropriate dataset, estimate a regression model, and present your model and results in a written academic essay (i.e. term paper). You can choose any topic or dataset that might interest you. If you need some inspiration, https://ourworldindata.org/ has fascinating long-run datasets, but you are welcome to use any data to answer any question you are interested in (for example: does smoking affect babies’ birthweight? Has the introduction of airbags/seatbelts lowered car accident deaths? Has the BC carbon tax lowered CO2 emissions? etc.).

Submit a half-page research proposal by February 13th, 4pm, as a PDF document, outlining the question you want to tackle and what data you are planning to use, including references. Your final project may deviate from this proposal.

Present your results in the form of a written academic essay (maximum length of 3 pages + references).

-     The project must be submitted via Brightspace by March 6th, 4pm, as a PDF document. Files submitted as Word document or Pages files will not be accepted and count as not submitted.

-     You must also submit your data and R-code used to estimate the models in your project via Brightspace.

The essay should be an academic piece of writing (avoid casual language, bullet points, make sure you correctly cite the relevant literature, and do not copy and paste R-code or output into the main text). The essay should follow a clear and coherent structure, for example you could structure your paper as:

1) Introduction

Stating your hypothesis, explain why this topic is important and relevant, what related literature already exists (correctly citing relevant papers), and how your work contributes to the existing literature on this topic.

2) Data

Describe your data (including sources and descriptions of variables), ideally including a well-labelled Figure. 3) Methods

Describe your estimated model (ideally as an equation), justify the functional form you choose (log, quadratic, etc.) and what assumptions you make in your model, and whether these assumptions are reasonable in your     context.

4) Results

Show your estimated model results in a Table including: coefficients, standard errors, number of observations, and R-squared (see below for an example in Table 1). Interpret your results carefully, conduct and interpret hypotheses tests (individual and perhaps joint tests). Do not copy and paste R-output or code (also no screenshots), but instead present your results in a clear manner so that someone not familiar with your data and code can understand it. Be careful about causal interpretations!

5) Conclusion

Summarise your results and discuss what you conclude based on your findings, consider shortcomings of your model and future research if you were to explore this question further.

6) References

References to the literature and data used.