1) Produce a scatterplot with the average wage at each education level on the y-axis, and education level on the x-axis. Discuss the relationship between wages and years of schooling based on this graph.

2) Generate a variable called edgrp that equals 0 for people who have a high school education or less (educ <=12) and equal to 1 for people who have more than a high school education (educ >12). Then plot a separate histogram for each group and describe the key differences.

3) Suppose you think that the model relating wages and years of schooling is = ! + " + Provide a precise interpretation of ! and " in the context of the model.

4) Describe some of the unobserved factors that are contained in the error term u.

5) Explain what the assumption [|] = 0 means in the context of this model, and explain fully whether you think it is true or untrue. What are the consequences for the OLS estimator 1 "if it is untrue?

6) Estimate the parameters of the model by OLS and interpret your estimates of ! and ". Note: use the robust option in your regression.

7) Compute the residuals from the regression in (6) and call them res. Then run another regression using res as the dependent variable and educ as the independent variable. Provide an explanation for why the slope in this regression equals 0 (hint: review the slide on the OLS Normal Equations).

8) Define the R2 and SER in the context of this model and use them to describe how well you think the estimated regression line fits the data.

9) At what significance levels would you reject the null hypothesis that one more year of education increases wages by $0.33/hour, in favour of the alternative that it does not?

10) Construct a 90% confidence interval for the change in wages from a half-year increase in years of schooling.

(THIS IS A STATA 13 ASSIGNMENT, I JUST NEED THE COMMANDS, NO CALCULATIONS, THANK YOU)