Question

Q1. For each part of this question, select one correct answer 3 points each). a. Statement: When estimating a regression with a binary dependent variable, it is necessary to use heteroseedasticity-robust standard error estimates to test hypotheses about the regression coefficients A Agree B. Disagree b. Statement: If MLR.5 is violated, we can still use the salt and F statistics when the sample size is large enough. A. Agree B. Disagree c. Statement: Suppose that the dependent variable in your regression has a no normal distribution, even after controlling for the variables (In other words, MLR.6 is violated) and therefore the OLS coefficients and standard errors cannot be used for hypothesis testing. A. Agree B. Disagree d. Statement: The linear probability model for a binary dependent variable necessarily has a heteroskedastic error term. A. Agro B. Disagree e. Consider the problem of estimating the effect of family income on college grade point average: colGPA = Bo + Bifamine + where actual family income famine") may deviate from reported family income faminc). To be specific, the measurement errore is additive, ie famine - famine" is assumed to be uncorrelated with famine ander What is the classical errors-in-variables (CEV) assumption in this model and is the estimator of Bi consistent in the model? A. Corfamine. c) = 0, B, is consistent. B. Cov(faminc,ca) = 0, B, is consistent. C. Cor(faminc, eu) = 0, is inconsistent. D. Cov(faminc, eu) = 0, $, is inconsistent

Answer 1

а) вяр b) Agree . c) Disagree d) Agree e) con/famine, &,)zo, ß, is consistent