Question

Final Exam (Open-book) Wrte ALL answers in your blue book. Marital status has two categories "Married" and "Single". Race has four categories White", "Black", "Hispanic" and "Asian". How many dummy variables will be needed in a regression if you want to categorize an individual based on marital status and race (separately, not jointly)? (Spts) A. Two. B. Three. C. Four. D. Five. E. Six. 1. Which of the following statement is true about heteroskedasticity? (Spts) 2. A. It affects the estimation of the parameters and the standard errors of the parameter estimates. B. It affects neither the estimation of the parameters nor the standard errors of the parameter estimates. C. It affects both linear regressions and non-linear regressions. D. None of the above. 3. Which of the following statement is true about Linear Probability Model? (5pts) A. It has a dummy on the left-hand side, and the dummy is interpreted as probability. B. It has dummies on the right-had side, and the dummies are interpreted as probability C. It can have dummies on both sides, and all the dummies are interpreted as probability. D. None of the above. 4. Which of the following statement is true about Linear Probability Model? (Spts) A. Since marginal effects are constant, the model is not widely used. B. Since marginal effects are not constant, the model is widely used. C. Although marginal effects are not constant, the model is still widely used. D. Since it is homoskedastic, it is a very useful model. E. Since it is heteroskedastic, it is not a very useful model. F. None of the above.

Answer 1

Ans1. Four explanatory dummy variable need to categories individual based on marital status and race, the regression model can be as given below,

Yi = a + b*Mi +c*D2i + d* D3i + e*D4i + u

Where Mi = 1 for married and Mi = 0 for single, D2i = 1 for White or D2i = 0 for other race, D3i = 1 for Black or D3i = 0 for other race and, D4i = 1 for Hispanic or D4i =0 for other race.

so, option(C) is correct.

Ans2. Option(D) is correct because if OLS is performed on heteroskedastic data then it does not affect on the unbiasedness estimation of the parameter but the variance of the estimated parameter will not be minimum, this means that estimate will not be BLUE estimators.  

Ans3. Option (A) is the correct option since LPM model has a dependent variable as the dummy variable which is on the left side of the model and it is interpreted as a probability.

Ans4. Option(C) is correct since marginal effects are not constant but it is still widely used because heteroskedasticity problem can be solved.