Can I use the fixed effects model as a generalization of the DID with multiple periods and multiple treatments?
Hello everyone. I am evaluating an annual school award program for achieving educational goals. I have a biannual panel with 5 points in time (9 years). I would like to get the ATT of each edition of the award on the grade of the schools in a standardized examination. I saw in a topics that could use a fixed effects model with time dummies including an interaction between an indicator of each treatment and a indicator of post treatment.
It would be something like this:
prog2008 = winner in 2008 and time >= 2008
...
prog2015 = winner in 2015 and time >= 2015
xtreg dependent i.year prog2008 ... prog2015, fe vce (robust)
Is that correct? Would I be getting ATT so? Does anyone know of any article that can serve as a reference for this method?
Thank you very much in advance!
Discussion thread, Mr. Woodridge # 15
Http://www.statalist.org/forums/foru...mmy-variables? P = 1293824 # post1293824
It would be something like this:
prog2008 = winner in 2008 and time >= 2008
...
prog2015 = winner in 2015 and time >= 2015
xtreg dependent i.year prog2008 ... prog2015, fe vce (robust)
Is that correct? Would I be getting ATT so? Does anyone know of any article that can serve as a reference for this method?
Thank you very much in advance!
Discussion thread, Mr. Woodridge # 15
Http://www.statalist.org/forums/foru...mmy-variables? P = 1293824 # post1293824
