Re: [R] cox goodness of fit

From: Roland Rau <roland.rproject_at_gmail.com>
Date: Thu, 31 May 2007 09:32:07 -0400

Hi,

assuming you are doing a Cox-PH-Model, you can check: library(survival)
?coxph.object

There it says that the components 'residuals' refers to the martingale residuals. I hope I recall it correctly but there exists a simple relation between the martingale residuals ('mgr') and the unmodified cox-snell residuals ('ucs'):
ucs = delta - mgr
where delta refers to your event indicator (0=censored, 1=event).

I don't have the books with me at the moment, but I think I learned about residuals in a survival context either in Klein/Moeschberger: Survival Analysis or in Tableman/Sung Kim: Survival Analysis Using S

I hope this helps,
Roland

Murray Pung wrote:
> Is there an implementation of the Cox-Snell residuals / Nelson-Aalen plot
> for goodness of fit?
>
> Or otherwise is there an appropriate Goodness of Fit diagnostic?
>
> Thanks
> Murray
>



R-help_at_stat.math.ethz.ch 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. Received on Thu 31 May 2007 - 14:05:17 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Thu 31 May 2007 - 14:31:24 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.