Dear all,
I am running a panel regression with random effects. I use industry- and time- dummies to control for the systematic differences across years and industries. I performed an equivalent OLS regression to obtain VIF values to check for multicollinearity. However, with both of these in the model there are slight multicollinearity concerns for one of my independent variables (VIF = 3.35).
My question is: Are the concerns big enough to drop one set of dummies from the model? And if so, how do I decide which one? The outcomes with only keeping one are almost the same, except that the industry-dummies increase the R2 a lot more.
Thanks for your help.