Multiple Choice
Finite-sample distributions of the OLS estimator and t-statistics are complicated, unless
A) the regressors are all normally distributed.
B) the regression errors are homoskedastic and normally distributed, conditional on X1,... Xn.
C) the Gauss-Markov Theorem applies.
D) the regressor is also endogenous.
Correct Answer:
Verified
Correct Answer:
Verified
Q39: Homoskedasticity means that<br>A)var(u<sub>i</sub>|X<sub>i</sub>)= <span class="ql-formula"
Q40: If the functional form of the conditional
Q41: (Requires Appendix material)If the Gauss-Markov conditions
Q42: Besides the Central Limit Theorem, the other
Q43: Slutsky's theorem combines the Law of Large
Q44: Estimation by WLS<br>A)although harder than OLS, will
Q46: The OLS estimator is a linear
Q47: E <span class="ql-formula" data-value="\left(\frac {
Q48: Under the five extended least squares
Q49: You need to adjust <span