Question

In multiple linear regression analysis, the number of independent variables should be

		A. more than 5.
		B. enough to guarantee that statistical significance is achieved.
		C. guided by economic theory.
		D. as large as possible.

2. Omitted variable bias is a problem because

		A. it causes the model to no longer be linear in the parameters.
		B. it prevents correctly estimating marginal effects.
		C. it prevents the model from being able to be estimated by ordinary least squares.
		D. it causes perfect multicollinearity.

Answer 1

1. As long as the independent variables are significant and the relationship between the dependent and independent variables is meaningful, it is fine. So, the number of independent variables should be enough to guarantee that statistical significance is achieved. Hence, Option (B) is the correct choice. (Ans).

2. Omitted variable bias occurs when a statistical model leaves out one or more independent variables, which could be relevant. It prevents correctly estimating the marginal effects of each variable. Hence, Option (B) is the correct choice. (Ans).