So your question is about fitting a regression model for all the subsets of predictors? Then there would be 2^13 submodesl? Probably leaps() does what you want. This function does a all-subset regresion. -- View this message in context: http://r.789695.n4.nabble.com/Regression-Models-tp4173278p4180447.html Sent from the R help mailing list archive at Nabble.com.