Obtaining hazard rate instead of cumulative hazard

[Hale Isaac]

Hi Statalist,

I am trying to produce that hazard rate at different time points. I cannot share my data due to it being sensitive data, but I can share the code I used.

I first set up my survival data

stset survtime, failure(fail==1)

I then get the cumulative hazard with:

sts list, cumhaz risktable (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14)

However, I would like to get the hazard rate not the cumulative hazard. I followed the advice here: https://www.stata.com/statalist/arch.../msg00658.html

ssc install stkerhaz
stcoz, estimate basech(H)
stkerhaz, basecha(H) bwidth(5) ci out (hazard_rate, replace)
use hazard_rate, clear

(stkerhaz (SSC) by Enzo Coviello)

I am not sure if this is the correct way to calculate the hazard rate? I am getting confused because I’ve seen suggestions to use this:

Sts graph, ci hazard
Sts graph, ci hazard outfile (hazard)

Saving the file gives me the estimates but no 95% confidence interval.