Which of the following is not a possible indicator of multicollinearity?
A) signs opposite from what is expected in the estimated $\beta$ parameters
B) non-significant $t$-tests for individual $\beta$ parameters when the $F$-test for overall model adequacy is significant
C) significant correlations between pairs of independent variables
D) non-random patterns in the plot of the residuals versus the fitted values

Question

Quizplus · Accepted Answer

Non-random patterns in the plot of the residuals versus the fitted values indicate issues with the model fit or heteroscedasticity, not multicollinearity. Multicollinearity concerns the correlation between independent variables, not the relationship between residuals and fitted values.

Which of the Following Is Not a Possible Indicator of Multicollinearity