[R] predict risk (% of death) at a certain time point

David Winsemius dwinsemius at comcast.net
Tue Dec 20 04:55:30 CET 2011


On Dec 19, 2011, at 10:05 PM, Bill Hyman wrote:

> Dear all,
>
> Does any one know how to calculate the risk (death rate) at some  
> time point (say 5 yrs) based on Kaplan-Meier curve in coxph model in  
> R? I know how to fit the survival data with coxph. But need to  
> calculate risk at some time point. Many thanks.

Have you looked at either:

?survfit.coxph
?predict.coxph   # with type ="expected"

This seems to be a particularly common question this week. Is there  
some sort of class project that is due before the holidays?


-- 
David Winsemius, MD
West Hartford, CT



More information about the R-help mailing list