[R] validation of logistic regression model with weighted presences and absences

[Frank E Harrell Jr]

Eva Mosner wrote:
> Hello everyone!
> 
> I want to validate logistic regression models of species distrubution
> data where presences and absences are weighted differently.
> I used lrm from Design package but unfortunately, validate.lrm ignores
> weights in the validation process.
> Has anyone an idea (package, code) how I can do the validation including
> weights?
> 
> Thanks a lot!
> 
> Eva
> 
> EvaMosner at web.de

Code contributions to the predab.resample function in Design are welcomed.

Frank

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