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
Related Posts with Propensity score matching
int(runiform(1,499999)) stata 13Hello to all, do you have a idea of wich command to use to have the same result... In stata 15 gen …
Regression tables : Dependent variable in rows with coeff in columns & mean of control groupHi, I need help with constructing regression tables which look like this: Panel A Control Group …
Obtaining 5 year rolling standard deviation of RETHello everyone: I am calculating the estimate of stock volatility, which use annualized standard de…
cgmreg - storing the number of observationsI have a problem in pubIishing the estimation tables using esttab as I am unable to store the statis…
How STATA handles missing dataGood evening, I have a relatively straight forward question regarding STATA. I have been trying to f…
Subscribe to:
Post Comments (Atom)
0 Response to Propensity score matching
Post a Comment