Homework 4 (4 Points), due at the beginning of class on September 26, 2016 Please use Wage data posted on my webpage to answer following questions. First, focus on the variables wage, dummy variable female, and another dummy variable married. • Q1 (1 point) Please use the command reg to (1) find the average wage for unmarried persons, and (2) report the two-sample t test for the null hypothesis that on average an unmarried person earns same wage as a married person. (Do not use command ttest!) • Q2 (2 point) Please interpret each β in the following regression wage = β0 + β1 female + β2 married + β3 (female ∗ married), where female ∗ married is the interaction term. Report the F test for the null hypothesis that the effect of gender on wage does not depend on the marital status. Please mathematically specify the null hypothesis. Run a restricted regression that imposes the null hypothesis. Show me the process of getting the F statistic. What is your conclusion? • Q3 (1 point) Focus on the variables wage, exper (working experience) and numdep (number of kids). Please test the hypothesis that whether having kids has no effect on the relationship between wage and exper. Hint: generate a dummy called kid using the following command gen kid = (numdep>0) This example shows you how to generate a dummy variable based on a numerical variable. An equivalent way to get the dummy is gen kid = 0 replace kid = 1 if numdep>0 1