Explain why or why not in each case. Which of the following are consequences of

Question

Explain why or why not in each case.

Which of the following are consequences of heteroscedasticity? A. The OLS estimators 0 are inconsistent. B. The usual F statistic no longer has an F distribution. C. The OLS estimators 0 are no longer BLUE. Explain why or why not in each case. Consider the following model based on cross-sectional data for questions 2 and 3. wage = constant + beta_1* education_i + beta_2 * experience_i + mu_i Where:| wage = the hourly wage for person i education = the number of years of education for person i experience = the number of years of working experience for person i

Explanation / Answer

1)

A) We cannot say they are inconsistent.

Inconsistency is due to correlation between error and regressors.

eg: To omitted variables, measurement error in the regressors, are but not to conditional heteroskedasticity.

B) Yes, it is no longer F distribution.

We need both A.MLR5 (conditional homoskedasticity) and A.MLR6 (errors conditionally normal) for the F-statistic to have a Fisher-F distribution under the null .

C) OLS are blue indicates they are best and efficient among all linear unbiased estimators. Hence the claim it is not BLUE is true

Related Questions

Navigate

• Browse (All)
• Browse E
• Subjects

---