Hi,
I know how to do propensity score modeling in stata with cross sectional data. But I dont know how to do it in stata when I have multiple years. Can someone please let me know?
Why do you want to use propensity-score matching?
Please explain more about your study design...ordinarily you would be matching on the "baseline" variables (even if you have longitudinal follow-up data) so I think an explanation of your question and how the data are structured is necessary...
As @Andrew D Althouse mentioned, for multiple years of data, you typically want to match using the first round of data (i.e., the "baseline" or pre-intervention data). This way, you will be able to calculate the DiD for your outcomes of interest (provided of course you're doing an impact eval of some kind).
As for the actual syntax, I'd suggest you look up the help file for "psmatch2" in Stata. Alternatively, Stata 13.0 or higher provides easier ways of doing through "tEffects".
Hope this helps. But if you have more details as to what you want to do, that will help.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling