[R] Survival questions

Stats Student @t@t@@@tudent4647 @end|ng |rom gm@||@com
Tue Jul 25 22:18:08 CEST 2023


Hi - I am using the survfit() function to produce Kaplan-Meier
survival curves for several different groups.

survfit (Surv() ~ cohort, data=d)

Everything works fine, but I'd like to do something different.

I have a 12 month survival curve (base) and I also have two survival
curves based on some experiment that only go out to 3 months. What I
would like to do is take the 3 month realized curves and apply the 3+
month hazard from the base curve to those.

1) is this a reasonable approach? (short of building a full-fledged model)
2) is there a standard way of doing this? I was going to read the
survival curves, calc the hazard (S(t) / S(t-1)), apply it to the
realized (short) curves. Related to that, from the survfit() object,
how do I read various curves specific to a particular cohort?
Everything appears to be concatenated together but I am guessing there
is some way to separate based on particular strata.

Thanks in advance.



More information about the R-help mailing list