Re: lrm

[Frank E Harrell Jr <f.harrell@vanderbilt.edu>]

jose Bartolomei wrote:
> Dear S-Plus users,
>
>
> I have a problem runing a logistic regression using the Design library.
>
> If I fit a lrm with more than two variables I obtain the following 
> result: (taking as an example the acath data.frame used in "An 
> Introduction to S and The Hmisc and Design Libraries")
> fit<-lrm(sigdz ~ age.tertile * sex + choleste)
> Problem in model.matrix.default(Terms.ns, X): No data for variable 
> "age.tertile"
> or if I use the rcs fuction
>
> fit<-lrm(sigdz~rcs(age,4) * (sex + rcs(choleste, 4)))
> Problem in model.matrix.default(Terms.ns, X): No data for variable 
> "rcs(age, 4)"
> What could be happening?
>
> Thanks,
> jose
>
>
> --------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu.  To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> the BODY of the message:  unsubscribe s-news
Hard to say.  You didn't following posting guidelines in sending enough 
of the code.  In particular I don't know if you attached a data frame or 
have all the variables already in the search path, and whether the 
tertiles were correctly calculated.
Frank

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