The data set "NLSW88.dta" on Courseworks contains information of a group of women in their 30s and early 40s which was used to to study labor force patterns. We will work with variables wage - a woman's hourly wage in 1988, tenure - the number of years the woman has worked with the current employer, and union - whether or not the woman belongs to a labor union. (a) (10 points) Run a regression of a woman's hourly wage in 1988 on her tenure, whether or not she belongs to a labor union, and the interaction term between these two variables. Show the STATA command and its output. Interpret the coefficients of this regression. (b) (10 points) Obtain the residuals of this regression and verify whether Assumptions 1, 5 and 6 of OLS regression are satisfied. Show the STATA commands and the corresponding graphs. What is your conclusion? (c) (8 points) Use a lincom command to calculate a 95% confidence interval for the hourly wages of a woman who has a tenure of 10 years and does not belong to a labor union. Show the STATA command and its output. What is the confidence interval? (d) (8 points) Use a lincom command to test if, after 10 of tenure, a woman who does not belong to a union makes no more than 1.5 dollars an hour less than a woman who belongs to a union. What would be the conclusion of the test using 5% significance level? Show the STATA command and its output, and justify your conclusion. Page 3