Hi,
I am using STATA to do propensity score matching on a large sample (n around 1000 or more).
Only way I know is the psmatch2 command, generate 1:1 matched pairs and identify them and run a univariable cox regression in those identical pairs. In the process I lose a lot of data.
I was wondering if anyone knows a way to use propensity scores as covriates or knows how to stratify the propensity score in blocks and use almost all data to run the regression.
Thank you in advance.
Dhaval
You can do it by specifying the 'neighbor' option. For example, 'neighbor(4)' means there will be four nearest neighbors (controls) for each case.
Hi, Md. Belal Hossain What syntax and command in Stata can I use for the nearest neighbors and how do I remove unmatched pairs?
Thanks in advance.
Dhaval
Use the following commands "psmatch2 exposure covariates, out(outcome) neighbor(4) caliper(0.01) ate". You should able to get the number of subjects in the treated and untreated groups.
To remove remove unmatched pairs for neighbor(1), you can try the following:
psmatch2 exposure covariates, out(outcome) neighbor(1) noreplace caliper(0.01) ate
gen pair = _id if _treated==0
replace pair = _n1 if _treated==1
bysort pair: egen paircount = count(pair)
drop if paircount !=2
For neighbor(4), you can use the following:
psmatch2 exposure covariates, out(outcome) neighbor(4) caliper(0.01) ate
gen pair1 = _id if _treated==0
replace pair1 = _n1 if _treated==1
gen pair2 = _id if _treated==0
replace pair2 = _n2 if _treated==1
gen pair3 = _id if _treated==0
replace pair3 = _n3 if _treated==1
gen pair4 = _id if _treated==0
replace pair4 = _n4 if _treated==1
bysort pair1: egen paircount1 = count(pair1)
bysort pair2: egen paircount2 = count(pair2)
bysort pair3: egen paircount3 = count(pair3)
bysort pair4: egen paircount4 = count(pair4)
egen byte paircount = anycount(paircount1 paircount2 paircount3 paircount4), values(2)
drop if paircount==0
I thank Dr. Ariel Linden, who initially wrote these commands.
I suggest you check these two websites too:
https://www.ssc.wisc.edu/sscc/pubs/stata_psmatch.htm,
https://rpubs.com/buidiengiau/psm-stata
All the best.
There are several ways to increase your sample size:
You can increase the caliper
Or omit some variables and accept wider bias correction
Or do 1:2 or higher match
Or use kernel ps
Or iptw
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics