[R] predicting residual expected survival times

[Dieter Menne]

Anne <anne.piotet <at> gmail.com> writes:

> Is there an implemented method to predict residual expected survival
> times for parametric/Cox PH models ? (I have modelled my data using
> the survival library)
> I would like to predict for a given subject (with a given profile )
> having survived up to time Ts the expected residual surviving time (or
> the  residual survival time quantiles)

What's wrong with documented resid?

 fit <- coxph(Surv(start, stop, event) ~ (age + surgery)* transplant,
               data=heart)
 mresid <- resid(fit, collapse=heart$id)

Dieter