1. If the regression model includes several categorical variables,is it rational to calculate VIF (Variance Inflation Factor) without converting them to dummy variables?
2. Is there a way to calculate VIF from variance-covariance matrix of regression coefficient without calculating statistics regarding predictors ? (I need this because Matlab regression gives cov(b) and handles categorical variables internally,but does not give the predictors variance. Please suppose variables are not mean centered)
Have you looked at Rapidminer? Open source tool with advanced modeling features.
Cheers
Sven
Dear Amir
you can get so much help about that in the book of Pidhazur 1997, it's a very good reference for regression analysis.
Rana
VIF is a measure of collinearity between two independent variables or
multicollinearity among three or more independent variables. It is the proportion of variance in one independent variable that is not explained by the remaining independent variables.. VIF values of just over 1.0 are desirable, with the floor VIF value being 1.0. A VIF value of 1.007 would be considered very good and indicative of no collinearity.
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series