-
@SethGershenson @shibettes @causalinf @Stata Three things - If there are other control variables, need to interact them with the group dummy too - Code I suggested is specifically for partly overlapping samples (think control vs. T1 vs. T2, and want to compare) - Should have cited @AustnNchols for original code reference!On twitter.com
♻️ 3 Retweets
❤️ 2 Favorites
Mood +1 🙂