[R] Estimating Survival function in coxph

Thomas Lumley thomas at biostat.washington.edu
Mon Oct 30 05:47:36 CET 2000

On Wed, 25 Oct 2000, Richards, Tom wrote:

> Hello:
> Assuming that I have used the coxph function to fit a Cox model with several
> covariates, is there a natural way to estimate a value of the survivorship
> function at, say, five years, 
> S(5), along with a confidence interval?  I have brought this question up
> with one or two colleagues, who have suggested using a parametric model, but
> I wonder whether it can be done using the semiparametric coxph() model.

Yes. The survfit() function does this. There are a few options for exactly
how to do it. They are more-or-less equivalent if a lot of people are
still at risk at this time.  It is necessary to estimate either the
baseline hazard function using something like the Nelson--Aalen estimator
or the baseline survival function using something like the Kaplan-Meier


Thomas Lumley
Assistant Professor, Biostatistics
University of Washington, Seattle

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list