Answer-first summary for fast verification
Answer: Include important variables that have been omitted and re-estimate the revised model
## Explanation When the error term is correlated with an independent variable, this violates the classical linear regression assumption of no correlation between regressors and the error term. This condition is known as **endogeneity**. Among the corrective actions: - **A. Use the natural logarithm** - This transformation may help with non-linearity or heteroskedasticity but doesn't address the fundamental issue of correlation between the error term and independent variables. - **B. Include important omitted variables** - This is the most appropriate action. When relevant variables are omitted from the model, their effects get absorbed into the error term, causing correlation between the error term and the included variables. Including these omitted variables removes this correlation. - **C. Use a subsample of data** - This might help with structural breaks or changing relationships but doesn't directly address the correlation between error terms and independent variables. **Correct Answer: B** - Including omitted variables is the primary solution to address endogeneity caused by omitted variable bias.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
Which of the following is the most appropriate corrective action if the error term in a regression model is correlated with one of the independent variables?
A
Use the natural logarithm of the independent variable in the revised model
B
Include important variables that have been omitted and re-estimate the revised model
C
Use the subsample of data most representative of conditions during the model forecasting period
No comments yet.