I have different treatments, on different groups, that occur at different times (e.g., 2000 on treat 1, 2009 on treat 2, etc.). I want to fit a DiD model, and I want to standardize the treatment years to year "0" as the the treatment time. To control for treatment year heterogeneity, I included the true year fixed effects.

\begin{align} y_{it} = &β_0 + β_1{\rm Treat} + β_2{\rm After} + β_3({\rm Treat} \times {\rm After}) + \\ &({\rm pre}_3 + {\rm pre}_2 + {\rm pre}_1 + {\rm now} + {\rm post}_1 + {\rm post}_2 + {\rm post}_3) + γC_{it} + ϵ_{it} \end{align}


  1. Is there anything wrong with this model? What should the time for the controlled group?

  2. Can I still include the time fixed effect as {pre1,2,3, now, post1,2,3}?

I know there is another way of writing the multiple treatment as to just include the interaction term (which do not require time standardization): variable("Treat") equals to 1 is the treated groups in the time when treated already, and variable("Treat") equals to 0 for controlled groups and for treated groups in the time when not yet treated. But I need the standardized method as well.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.