Explanation:
Regression equations are based on the principle that there is an explaining part, which are the independent or explanatory variables in the equation, and an unexplained part, which is the error or residual component of the regression. When variables useful in explaining the behavior of the dependent variable are omitted from the regression formula, their effect is captured in the error, or residual component. This causes the error to no longer be random from the dependent variable, which negates one of the underlying assumptions of OLS analysis. The regression estimates are biased, and hence the regression results are wrong.
(Book 2, Module 20.2, LO 20.d)
Ultimate access to all questions.
Question 33
A junior analyst is having difficulty with the concept of omitted variables in a regression. She recognizes there is always a potential for the explanatory portion of a regression to be incomplete, but she does not know exactly how to explain the potential impacts. Which of the following statements best describes the impact of omitting variables in a regression?
A
Omitted variables induce heteroskedasticity in the regression equation.
B
Omitted variables induce multicollinearity in the regression equation.
C
Omitted variables increase the F-test of the regression equation.
D
Omitted variables may help explain dependent variable behavior, which may lead to biased estimates.
No comments yet.