[R] stepAIC + cox.zph

Eric Rescorla ekr at rtfm.com
Mon May 11 23:06:38 CEST 2009


I'm using stepAIC with a CoxPH model on some data. As
we expect, there are a bunch of predictors, and stepAIC
throws some out and then converges on others.

Anyway, when I run cox.zph on the final model, I get
some evidence of Beta(t) time dependence. Should I
be concerned about the impact of this on the entire
fit procedure, or is it just that I can't trust the final
hazard ratio that the Cox model spits out?

Thanks in advance,

More information about the R-help mailing list