Economics 391 (Spring 2013) Professor Lamarche, University of Kentucky
Sample Final Exam Questions
1. Suppose you estimate a model obtaining,
wage = 0.620 + 0.506 × educ + 2.27 × female
were wage is measured in dollars/hour and female is a binary variable that is equal to one if the person is a female and zero if the person is a male. The standard errors are 0.67, 0.050 and 0.279, respectively.
(a) Is the ‘gender gap’ significantly different from zero? (b) What is the predicted wage for a female worker with 10 years of education with 16 years? (c) How many years of education a male needs to obtain the hourly wage that a female with 16 years earns?
2. Consider you estimate the following regression model,
y = a + ßx + u
where y is the dependent variable, x is the independent variable and u is the error term.
(a) Explain the difference between correlation and regression and explain the assumptions of the regression model. (b) What are the interpretation of a and ß? (c) How would you estimate the unknown population parameters? (d) What are the statistical properties of the method you proposed?
3. Suppose that the null hypothesis is H0 : = 1000 and the alternative hypothesis H1 : =? 1000. Assuming s = 200, n = 100, X¯ = 980, and a = 0.01, calculate
Document Preview:
Economics 391 (Spring 2013) Professor Lamarche, University of Kentucky Sample Final Exam Questions 1. Suppose you estimate a model obtaining, wage= 0.620+0.506×educ+2.27×female were wage is measured in dollars/hour and female is a binary variable that is equal to one if the person is a female and zero if the person is a male. The standard errors are 0.67, 0.050 and 0.279, respectively. (a) Is the ‘gender gap’ signi?cantly di?erent from zero? (b) What is the predicted wage for a female worker with 10 years of education with 16 years? (c) How many years of education a male needs to obtain the hourly wage that a female with 16 years earns? 2. Consider you estimate the following regression model, y = a+ßx+u where y is the dependent variable, x is the independent variable and u is the error term. (a) Explain the di?erence between correlation and regression and explain the assumptions of the regression model. (b) What are the interpretation of a and ß? (c) How would you estimate the unknown population parameters? (d) What are the statistical properties of the method you proposed? 3. SupposethatthenullhypothesisisH : µ = 1000andthealternativehypothesis 0 ¯ H : µ = 6 1000. Assuming s = 200, n = 100, X = 980, and a = 0.01, calculate 1Economics 391 (Spring 2013) Professor Lamarche, University of Kentucky the value of the test statistic, set up the rejection region, and determine the p-value. Would you reject the null hypothesis? 4. Consider the return to education model, lnw =ß +ß educ+u it 0 1 where the dependent variable is the logarithm of wages and the independent variables are years of education (x ) and years of experience (x ). 1 2 (a) Suggest a method to estimate the equation of interest. (b) Using the method you proposed before, you obtain the regression results called “SUMMARY OUTPUT # 1” (please see the last pages). What is ˆ ˆ the interpretation of the coe?cients ß and ß ? 0 1 (c) Test if education has a signi?cant e?ect on log of wages. (State clearly the null...