(f) Estimate the regression of GPA on age, sophomore, junior and senior for men only. What happens to the statistical significance of the coefficients in this regression? Explain why this is the case (g) Estimate the same regression as in part (c), but now do this for all respondents (male and female). Include controls for male, work, marijuana, lightdrinker, moddrinker and heavydrinker. Calculate the predicted GPA for a male senior who works, is a moderate drinker and has not smoked marijuana in the past 30 days. (h) Using the regression in part (g), test the hypothesis (at the 5% significance level) that freshmen and sophomores have the same GPA on average, holding constant gender, work, type of drinking and marijuana use. (i) What is the adjusted R2 for the regression in part (g)? What does this adjusted R2 tell you about the fit of the regression? Does it indicate that omitted variable bias is likely to be a problem? 5. Stock and Watson Empirical Exercise E 5.3, parts (a), (b) and (c), and Stock and Watson Empirical Exercise E 6.1 part (b) questions (i) and (ii) only. For these exercises, use the birthweight_smoking. cav dataset that is posted on Latte. 6. Stock and Watson Empirical Exercise E 7.1, parts (a), (c), (d) and (e) only 7. True False Uncertain questions. Explain fully your answers. a. Suppose we have the following model: ?; = Bo+ B1?; + B2?? + ?; The effect on Y of a change in X will depend on the level of X. b. Suppose ?ar(? ) is positively correlated with X in the population. In this case, the OLS estimators are biased (Unless otherwise stated, assume all relevant assumptions hold) c. To test whether or not the population regression function is linear rather than a polynomial of order y, one should use the test of (7-1) restrictions using the F-statistic. d. If By is significantly different from zero at the 5% significance level, then X has a causal impact on Y. e. In(?) = B1 + 2 In(?) + ? is linear in both the parameters and the variables. f. If the true model is ? = By + By? , + 83?2 + ? but you estimate ? = By + B2? 1 + ?: your estimate of B2 will always be biased