[R] prevalence in logistic regression lrm()

Harrell, Frank E f.harrell at Vanderbilt.Edu
Sat Jul 11 17:42:02 CEST 2009


This is dangerous because prevalence is an ill-defined concept.  Most
discussions of prevalence fail to understand conditioning completely.

If you must, you can do this with an appropriate offset() term.

Frank


>
> Hi, I am wondering if there is a way to specify the prevalence of events
> in logistic regression using lrm() from Design package? Linear
> Discriminant Analysis using lda() from MASS library has an argument
> "prior=" that we can use to specify the prevalent of events when the
> actual dataset being analyzed does not have a representative prevalence.
> How can we incorporate this information in lrm()? The concern I have is if
> my dataset does not have a representative prevalence, then the probability
> generated by lrm() will not be meaningful. Thanks
>
> John
>
> ______________________________________________
> R-help at 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.
>




More information about the R-help mailing list