lrberge / fixest

Fixed-effects estimations
https://lrberge.github.io/fixest/
362 stars 59 forks source link

Multiple Treatment Types in Staggered (`sunab()`) #382

Open mikenguyen13 opened 1 year ago

mikenguyen13 commented 1 year ago

Hi, I am trying to implement the package to the setting of multiple treatment types (e.g., a firm can experience treatment 1, treatment 2, treatment 3). And each unit can have a different treatment combination. I was wondering how I could apply your package in this setting. Since sunab() cannot take 2 cohorts (I was trying to specify cohort within each treatment type, and treatment type as another cohort), I thought that I could include the treatment type as the fixed effect. But in this case, it does not allow for the spillover effect of one treatment type on another. I was wondering if you could suggest a way to apply your code to the multiple treatment types setting. I really appreciate your time.

kylebutts commented 1 year ago

@mikenguyen13 I think that you're right that sunab() can't take multiple cohorts at once. If you're willing to make assumptions (parallel trends for all combinations of treatment and additive separability of each treatment from one another) and have a never-treated group, you can use https://github.com/kylebutts/did2s and include each treatment in the second_stage.