Answer-first summary for fast verification
Answer: The standard error estimates for the slope coefficients are too small.
## Explanation **B is correct.** The presence of heteroskedasticity affects the efficiency and the accuracy of the standard errors of the estimates, but not the unbiasedness of the coefficients. Specifically: - Heteroskedasticity causes the usual OLS standard errors to be biased and often leads to underestimation - This affects hypothesis tests and confidence intervals, making them unreliable - Test statistics (such as t-tests) may be inflated, leading to a higher likelihood of Type I errors (incorrectly rejecting the null hypothesis) **A is incorrect.** Heteroskedasticity does not affect the consistency or the unbiasedness of the OLS parameter estimator. If heteroskedasticity is present in a regression, the estimates for the slope coefficients themselves are not biased. **C is incorrect.** The explained sum of squares (ESS) and residual sum of squares (RSS) follow a specific relationship with the total sum of squares (TSS): TSS = ESS + RSS. If you divide the ESS by the TSS, you get R², which represents the proportion of the variance in the dependent variable explained by the regressors. This R² value will be between 0 and 1 but does not imply that ESS and RSS individually sum to 1. **D is incorrect.** Heteroskedasticity indicates that the standard deviation of the residuals is widening as the regression line goes further out, not that they are too low.
Author: LeetQuiz .
Ultimate access to all questions.
No comments yet.
A researcher is running a regression to test the hypothesis that the forward interest rate is an unbiased predictor of the future spot rate. The researcher uses a large sample size but is concerned about the potential impact of heteroskedasticity on the estimates of the ordinary least squares parameters. If the regression residuals exhibit heteroskedasticity, which of the following observations is most likely for the researcher to make?
A
The coefficient estimates will be biased.
B
The standard error estimates for the slope coefficients are too small.
C
The explained sum of squares will be equal to 1.
D
The standard deviations of the residuals are too low.