Diff in Diff, post vs date fixed effect

[Saharsh Agarwal]

Hi,

I have 2 groups (1 treated), and data at the daily level. The treatment occurs on a specific day.
I estimate the following 2 DD model:

$y_{it} = \alpha treated_i + \beta post_t + \gamma treated_i \times post_t + \epsilon_{it}$
where I refers to the groups. and post is 1 for all periods after the treatment occurs.

&

$y_{it} = \alpha_i + \beta_t + \gamma treated_i \times post_t + \epsilon_{it}$

The only difference between the two models is that I have a fixed effect for each day, rather than the just pre and post. I am getting the same DD estimate for both (not sure why), but a senior says they shouldn't be the same. Can someone please help resolve this?

Thanks!

[Clyde Schechter]

I think you need to show the actual code and Stata output to get a useful response. If you are not familiar with using code delimiters to do that, please read the Forum FAQ, with special attention to #12.