Answer Happy

PART 3 (Questions 13-20) (34 points) We are investigating the relation between wages and education. To test the effect of education on wages of workers, we specify the model wage Bo + Bimale + B2south + Bawhite + Bieduc+u, where wage is the hourly wage in U.S. dollars, educ denote the years of formal education, and male, south and white are du variables. male equals 1 for males and is zero for females, while south equals 1 if the worker is from the South of the cour zero if the worker is from the North. Similarly, white equals 1 for white, O for other ethnicities. Question 13

Question 13 Which fitted line describes white females from the South? O wage - Bo + B₁ + Beduc O wage Bo+(1-B₁) + B₁ + Beduc wage=Bo + B₁ + Beduc wage=Bo + B₂+ Beduc O wage=Bo + B₂+By+Beduc

A 20-4-Agde OE-A₂+A₁+₂+deduc Question 15 Which fitted line describes non-white male from the North? O wage=B+B₁ + Biedue wage - Bo+B₂+ Beduc wage=B+Beduc O wage=B+B₁ + Beduc wage=B+B₁ + B₁ + Beduc 3 pts

0 0 Question 14 Which fitted line describes non-white females from the North? Ouage=A+B₁ + Beduc O wage= Bo + Breduc wage=Bo + B₂+ Beduc wage=Bo + B₂ + B + B₁educ wage=Bo + B₁ + B₂ + B₁educ Question 15 3

D Ⓒd₂+ A₂ + Ac O4 + A₁-A₂+ Ad Question 16 In a plot of education (in the x-axis) and wages (in the y-axis), suppose that the fitted line for white males from the North is above the fitted line for non-white males from the South. What is the sign of B₂? Positive Negative Not enough information provided 5 pts Question 17

O Not enough information provided Question 17 In a plot of education (in the x-axis) and wages (in the y-axis), suppose that the fitted line for white males from the North is above the fitted line for non-white males from the North. What is the sign of B? Positive O Negative Not enough information provided Question 18 5 pts Suppose that 3, is positive, check all the statement below that are correct:

Question 18 5 pts Suppose that is positive, check all the statement below that are correct: Holding education constant, a non-white female from the North is expected to receive a higher wage than a non-white male from the North. Holding education constant, a white female from the South is expected to receive a lower wage than a white male from the South Holding education constant, a white female from the South is expected to receive a lower wage than a white male from the North. O Holding education constant, a non-white female from the North is expected to receive a lower wage than a non-white male from the Sou Not enough information provided. Question 19 What is the new econometric specificatic

Question 19 Now you would like to test whether the educ effect varies by ethnicity. What is the new econometric specification you will have to run? What is the null hypothesis? Ouage-A + male+South+white+ Seduc + south educ + u H₂:1-0 O wage - Ag+male+south+white+ Baeduc + Bueduc white+u He: A-0 O wage = By +3male +Bsouth+Bawhite + Bieduc+ Beduc male + u Ha:-0 5 pts Owage=A+Amale+south+Bywhite + Beduc + Bseduc white +u Ho: B=B=B₁ = 0 O wage=B+Bimale + Basouth+Bawhite + Beduc + Bseduc male + u Ho: B₁=B4=Bs=0 O wage Bo + Bimale + B₂south + Bawhite + Beduc + Bseduc south+u Ho: B₂=B4=B₁ = 0

Question 20 One of your classmates suggests that you run the following specification: wage= = Bo + Bimale + B₂ south + Bawhite + Bieduc +Bs female educ + Benorth educ + Binonwhite educ + u - Note that for every individual it must always hold that male + female 1, north + south = 1, and white + run this specification? No, because the model is not linear. No, because it has a problem of perfect multicollinearity. O No, because the independent variables are correlated with the error term. O Yes, we can run this specification.