Dear R-users,
I am looking for a function designed to handle parametric proportional hazard
models with a piecewise constant baseline hazard (i.e. dummies for annual
intervals) and time-dependent covariates since I'm especially interested
about the effect of those covariates on the baseline hazard.
I tried survreg() but this doesn't seem to work since my data looks like
this
id/ start/ stop/ censoring/ covariates
1/ 0/ 5/ 0/ ...
1/ 5/ 8/ 1/
2/ 0/ 4/ 0/
and survreg(Surv(start,stop,censoring)~covariates,...) always returns
'invalid survival type' (specifying the type="right" returns
"wrong number of args...").
Am I missing something or is there a function that can handle this kind of data.
Thanks
Daniel
--