PROC PHREG with IPTW weighted cohort: robust standard error

dac_js — Fri, 08 Dec 2023 21:08:28 GMT

I have the following data:

data have;
input ID $ Time_to_Event Event Age Treatment $ Weight;
datalines;
01 12 1 45 A 1.2
02 34 0 50 B 1.0
03 25 1 60 A 1.5
04 44 0 55 B 0.8
05 15 1 65 A 1.1
06 8 0 40 B 0.9
07 20 1 70 A 1.4
08 30 0 35 B 1.3
09 22 1 42 A 1.0
10 18 0 57 B 1.2
11 36 1 48 A 1.1
12 40 0 53 B 0.7
13 5 1 62 A 1.6
14 28 0 37 B 1.5
15 16 1 44 A 0.8
16 45 0 56 B 1.0
17 9 1 39 A 1.3
18 27 0 51 B 1.2
19 33 1 64 A 0.9
20 11 0 46 B 1.4
;
run;

proc phreg data=have;
class Treatment;
model Time_to_Event*Event(0) = Age Treatment;
weight Weight;
run;

Is there a way to add robust standard error here to account for the weight added in the Cox regression?

Re: PROC PHREG with IPTW weighted cohort: robust standard error

OsoGris — Mon, 11 Dec 2023 11:26:16 GMT

Robust standard errors are produced by the COVSANDWICH(AGGREGATE) option on the PROC PHREG statement. You will also need an ID statement to list the variable that defines your clusters.

topic PROC PHREG with IPTW weighted cohort: robust standard error in Statistical Procedures

PROC PHREG with IPTW weighted cohort: robust standard error

Re: PROC PHREG with IPTW weighted cohort: robust standard error