[R] coxph with time-dependent covariates

Terry Therneau therneau at mayo.edu
Tue Jan 14 13:43:02 CET 2014

> R throws out this event due to the missing time-dependent covariate at day
> 6. Is there a way I can keep this event without filling in a covariate
> value at day 6?


The Cox model assesses the risk associated with each covariate by comparing, at each event 
time, the values of the subject who had the event to all those who were also present at 
that time.  You can't compare to missing.

Terry T.

More information about the R-help mailing list