Hi all,
I am currently trying to run PSM on STATA14 to compare effect of treatment 1 vs treatment 2 (control). I have a few questions before attempting the analysis:
1. How do I start the matching? Or am I right to say the teffects command would match treatment and control groups?
2. To work out the standardised difference, I presume I have to work out the means for each covariates prior to matching and then run the pstest command to obtain the means and standardised bias/ difference post-matching?
3. After analysis for example with teffect, unmatched group is displayed. However, I am more interested in the control group. Are they the same? If not how do I gather that?
4. How do I obtain the number of observations in both treatment and control groups post-matching?
Sorry for all the questions. I am relatively new to STATA so would appreciate if you guys can help guide me with this.
Regards
Ken
0 Response to Propensity score matching
Post a Comment