Question

1. A 2013 study in the USA investigated the relationship between employees’ wages, gender, race, union membership, education and work experience.
Equation 1 is the least squares estimated model (standard errors in parentheses):

Equation one:
?????= −7.183 −3.08?1? −1.556?2? +1.116?3? +1.371?????????? + 0.166???????????
       se= (1.015)   (0.364)     (0.509)        (0.508)         (0.066)    (0.016)

R2=0.234, n=1,289

Where Wage is measured in thousands of dollars; D1i=1 for Female, 0 for male; D2i=1 for Non-white, 0 for white; D3i=if union member, 0 for non-union member; education is number of years of education and experience is number of years of work experience.

a. Interpret the regression results from equation one. Conduct appropriate t and F tests for the significance of all independent variables in the model at 95% level of significance.

b. Explain what is heteroskedasticity and outline the consequences of heteroskedasticity.

c. How can we detect heteroskedasticity?

d. What are the remedies for heteroskedasticity?

e. Consider the below test results for heteroskedasticity in equation one. Interpret these results.

Heteroskedasticity Test: White
F-Statistic 4.898           Prob. F (17,1271) 0.000
Obs*R-squared 79.257               Prob. Chi-Square (17) 0.000
Scaled Explained SS 366.027   Prob. Chi-Square (17) 0.000

Answer 1