I am trying to calculate the number of observations within a time window. Specifically, for 1 hour before and 1 hour after the arrival time. I tried it for 20 observations, and the code runs properly, but my dataset consists of 250,000 obs and this loop has taken >1.5 hours to run. Would appreciate suggestions for a more efficient method.
Currently I have:
local N=_N
quietly forval i = 1/`N' {
count if (arrivaltime >= arrivaltime[`i']-1 & arrivaltime<=arrivaltime[`i']+1)
replace countcall = r(N) in `i'
}
My data looks like this:
PatientID ArrivalTime
1 2015-02-17 19:35:00
2 2015-02-17 20:42:00
3 2015-02-17 21:50:00
Already tried an egen command but it didn't work.
Aucun commentaire:
Enregistrer un commentaire