[R] rcspline.problem

[Frank Harrell]

Please follow the posting guide.  You didn't state which package you are
using and didn't include a trivial self-reproducing example that causes the
error.  

For your purpose the rms package is going to plot restricted cubic spline
fits (and shaded confidence bands) more flexibly.

Frank


Haleh Ghaem Maralani wrote:
> 
> Dear Dr ;
> 
> 
>    I am a PhD student at Epidemiology department of National University of
>    Singapore. I used R command (rcspline.plot) for plotting restricted
> cubic
>    spline  –  the  model is based on Cox. I managed to get a plot
> without
>    adjustment  for  other  covariates,  but I have a problem regarding to
>    adjusting the confounders.
> 
>    I applied below command to generate the matrix for covariates.
> 
>    m=as.matrix(age,sex)  or  m1=matrix(age,sex)  or m2=cbind(age,sex)
> 
>    But, when I input  ..... adj=m, or adj=m1, or adj=m2......  in the
> model, R
>    gives below error:
> 
> 
>    Error in pchisq(q, df, lower.tail, log.p) :
> 
>      Non-numeric argument to mathematical function
> 
>    In addition: Warning message:
> 
>    In coxph.fit(cbind(x, xx, adj), cbind(y, event), strata = NULL,  :
> 
>    Loglik converged before variable  1,2,3,4 ; beta may be infinite.
> 
> 
>    I would be grateful if you take my issue into your consideration and
> help me
>    on this case
> 
> 
>    Sincerely Yours
> 
> 
>    Haleh Ghaem
> 
>    PhD student, NUS
> ______________________________________________
> 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.
> 


-----
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context: http://r.789695.n4.nabble.com/rcspline-problem-tp3501627p3502623.html
Sent from the R help mailing list archive at Nabble.com.