Dear R-listers, I am building a prediction model starting with many, many variables. I use the 'stepAIC' procedure in the MASS package, and the model building process takes several hours to complete. At the very end, I am occasionally confronted with warnings like this one:
Warning messages: 1: In fitter(X, Y, strata = Strata, offset = offset, weights = weights, : Loglik converged before variable 4 ; beta may be infinite. In these instances, I take it that I have stretched my data too thin, and I discard the final model. Therefore, I would save valuable time if the procedure would stop automatically when it encounters any anomaly that causes it to warn me. Is this possible? And if so, how? Thank you in advance for any assistance. Peter Jepsen, MD ---- > sessionInfo() R version 2.8.1 (2008-12-22) i386-pc-mingw32 locale: LC_COLLATE=Danish_Denmark.1252;LC_CTYPE=Danish_Denmark.1252;LC_MONETARY= Danish_Denmark.1252;LC_NUMERIC=C;LC_TIME=Danish_Denmark.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] MASS_7.2-45 > ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.