[R] expected survival from a frailty cox model using survfit

Emmanuel Tillard tillard at cirad.fr
Wed Aug 2 17:13:36 CEST 2006

Hello R users

Would somebody know how to estimate survival from a frailty cox model, 
using the function survfit
and the argument newdata ? (or from any other way that could provide 
individual expected survival
with standard error); Is the problem related to how the random term is 
included in newdata ?

kfitm1 <- coxph(Surv(time,status) ~ age + sex + disease + frailty(id, 
dist='gauss'), kidney)
survfit(kfitm1) #ok for mean expected survival

Call: survfit.coxph(object = kfitm1)

      n  events  median 0.95LCL 0.95UCL
     76      58      63      39     132

survfit(kfitm1, newdata=kidney[1,]) #return an error message

Erreur dans x2 %*% coef : arguments inadéquats

Thanks in Advance

Emmanuel Tillard
Unite de recherche 18

UMR868 Elevage des Ruminants en Regions Chaudes (ERRC)
2 place Viala
34060 Montpellier cedex 1

tel:	0499612265 (fixe)
	0633850598 (gsm)
fax:	0467545694
e-mail: tillard at cirad.fr

More information about the R-help mailing list