The regression below captures the compensation (in $100000) among C-Suite executives. Model I: CEOWage (with hat on top) = 5.48 - .34female, where female is a binary variable that takes the value 1 if female, 0 if male. The standard error of B subscript 0 (with hat on top) is .02 and B subscript 1 (with hat on top) is .06. The total no of observations is 1200.
and R squared = .15.
I. Conduct hypothesis testing to show that female C-Suite executives earn less than male counterparts at 5% level. Show all the necessary steps.
II. Suppose you want to use the same data but would like to regress CEOWage on Male, a variable that is equal to 1 if the person is male and 0 if the person is female. Give the regression output for this regression.
III. Now consider, Model II: CEOWage (with hat on top) = 3.16 - .29female + .37firmsize + .0004 returnonequity + .98exper
The R squared for Model II is .38. Compare the estimated wage gap under different specifications of Model I and Model II. Using the right conceptual ideas explain the difference. Out of Model I and Model II, which model you would choose for the estimation?