[R] logistic regression model validation through bootstrapping

Frank E Harrell Jr f.harrell at vanderbilt.edu
Wed Sep 24 18:50:48 CEST 2008


Van Dyke, Alison wrote:
> I'm new to R and to validation of logistic regression models via bootstrapping.  I see that there are various approaches to bootstrapping, and I am wondering (1) which approach would be best for internal validation of predictive discrimination, (2) are there any good resources that could be recommended to explain the optimal approach, and (3) what packages in R should I use to achieve this approach.  Many thanks for the input.
>  
> Alison Van Dyke
> MD/PhD Candidate, Class of 2011
> Cancer Biology
> Wayne State University
> avandyk at med.wayne.edu
>  
> 

One possibility:
library(Design)
?validate.lrm

-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University



More information about the R-help mailing list