Question 4 1. We saw in class how to create a dummy variable that takes on values of 1 and 0. We also saw how we can use multiple dummy variables when we have more than two categories. Another important historical event in Africa's past was colonial rule, and we might be concerned, for example, that wealthier areas of Africa were more likely to engage in the slave trade and because of their wealth were more desirable targets for colonisation. So, maybe we would like to account for colonial rule. Add the variables colonyi where i = 1,..., 7. That is, use the OLS to estimate the following multiple regression model: gdpper capita, o + Bslaveexports; + B2coastline + Bpopdensity+ B4colony1B5colony2+ colony3+ Brcolony4+ Bscolony5+ Bocolony6+ B10colony7+ ui (6) Present your results in a table like those we've seen in class, be sure to use heteroskedastic- robust standard errors. 2. Why did I leave out colony in the model above? What would happen if you included it? 3. How many degrees of freedom do we have now? 4. After accounting for a country's colonial history, does the effect of slave exports remain statistically significant at the 5% level? 5. Conduct a joint hypothesis test on all of the colony variables, are they jointly signif- icant at the 5% level? (Hint: Don't forget to use heteroskedastic-robust test statistic.)