From the teffects documentation in Stata 14, I see that the margins command may be used to calculate the conditional ATE or ATET for a subpopulation for etpoisson and etregress.
Both etpoisson and etregress also take complex survey weights (using svy, pweight, etc.). However, the margins post-estimation command does not. It seems like my options are to re-parameterize etregress as needed, or ignore the biased standard errors.
Specifically, I want to model a treatment - covariate interaction (treatment * income). The example on page 49 of the documentation shows how to calculate the interaction, but does not use the complex survey weights.
Hello Richard,
That's odd - running the margins postestimation example after etpoisson with the svy prefix works fine for me in Stata 14.1 (see help etpoisson documentation).
Anyway, instead of using the svy prefix you might consider using the pweight and vce(cluster) options when specifying etpoisson. This will yield exactly the same results and margins should work in any case.
Regards,
Alex
I regret that I was not clear. I was having trouble with linearized Taylor series standard errors. Using vce(cluster) is less conservative that linearized Taylor series standard errors, but the estimates are the same, as you point out. However, when I re-ran the code today, margins after etregress worked just fine and calculated linearized s.e. I must have had a typo!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software