Can I drop year the time period indicators when I include year fixed effects in the model?

[Shuting Wu]

In a DID model, y = a + b*post + c*treat + d*post*treat + e, can I drop the variable post (an indicator that equals 1 for the time period after the regulatory change and 0 otherwise) if I also include year fixed effects in the model? (note the post period is not fully aligned with calendar year, the regulatory change does not happen on 1/1). Thank you.

[Carlo Lazzaro]

Shuting:
welcome to this forum.
Please read the FAQ on how to post (more) effectively, posting within CODE delimiters what you typed and what Stata gave you back. Thanks.
That said, have you experienced collinearity issues that led you to the approach you're intended to follow?

[Maxence Morlet]

Shuting Wu do you have panel data? If so, I would include unit and time fixed effects if I were you, which would cause treat and post dummies to drop.

You can also follow recommendations in Wooldridge's (2021) amazing paper: https://economics.princeton.edu/wp-c...Wooldridge.pdf