6 Exercises
Download the traffic fatality dataset used in Stock and Watson examples. It covers data from 48 states observed over seven years (from 1982 to 1988), for a total of 336 observations.
| Variable | Description |
|---|---|
| mrall | Vehicle Fatality Rate (VFR) |
| beertax | Tax on Case of Beer |
| mlda | Minimum Legal Drinking Age |
| jaild | Mandatory Jail Sentence |
| vmiles | Average Mile per Driver |
| unrate | Unemployment Rate |
| perinc | Per Capita Personal Income |
- Estimate a model for Vehicle Fatality Rate using your choice of variables listed above.
- Estimate a fixed effect model and test for state and time fixed effects.
- Run the necessary tests to check whether state and time fixed effects are present.
- Estimate a twoway model and compare to the previous state and time fixed effect models.
- Test for serial correlation and cross sectional dependence in the twoway model.
- If either serial correlation or cross sectional dependence is present, use the methods you’ve learned to obtain heteroskedastic and autocorrelation consistent standard errors.
- Compare the HAC and spatially robust standard errors with the twoway model estimated earlier.
- Display the results in publication-ready tables and discuss the substantively significant findings.