Hi all, I'm using the mle.stepwise function in the wle package to run forward selection. My statistical background tells me that the first variable selected should be the one with the highest correlation with the response, however that's not the case. The two highest correlations with the response are "similar" (0.86 and 0.89) however the forward selection selects the variable with 0.86 correlation first and then in the next step chooses the variable with correlation 0.89. Any thoughts on this? Is is this some type of bug in the function or package? Thanks -- View this message in context: http://old.nabble.com/Forward-selection-peculiarity-tp26260670p26260670.html Sent from the R help mailing list archive at Nabble.com.