Multicollinearity
Posted: Mon Oct 13, 2014 2:22 pm
Two of the explanatory variables in my linear regression are correlated (0.75). I need to include the two variables in my model. Judge et al. (Introduction to the theory and practice of econometrics) mentions about two solutions: (1) additional sample, and (2) Ridge Regression. RATS Manual for Additional Topics include some Ridge Regression programing, but I am not sure how to program the choice of K. What is the best way to handle this? Thank you.