Issue with different event-study functions

[Chenghao163]

I shows a event-study did example:
y = a + event_1treatbeta1+ event_2treatbeta2 + event_3treatbeta3 + event_4treatbeta4 + event_5treatbeta5 + u
In this function, using this code in stata may be efficient: honestdid, pre(1/3) post(3/4) mvec(0.5(0.5)2)

Instead, facing a similar gruop of functions
y = a + event_1treatbeta1 + u
y = a + event_2treatbeta2 + u
y = a + event_3treatbeta3 + u
y = a + event_4treatbeta4 + u
y = a + event_5treatbeta5 + u
How could I using honestdid code to caculate whether trentment violation of parallel trends?

[Chenghao163]

The plus signal is ignored, the real function is:
y = a + event_1×treat×beta1+ event_2×treat×beta2 + event_3×treat×beta3 + event_4×treat×beta4 + event_5×treat×beta5 + u

y = a + event_1×treat×beta1 + u
y = a + event_2×treat×beta2 + u
y = a + event_3×treat×beta3 + u
y = a + event_4×treat×beta4 + u
y = a + event_5×treat×beta5 + u

[jonathandroth]

Is your question how to use the package if each of your event study coefficients comes from a different regression? You should be able to use sureg and pass the vector of coefficients and vcv to HonestDiD. Or you can do it all yourself in a stacked regression

…

On Thu, Nov 14, 2024, 9:12 AM Chenghao163 ***@***.***> wrote: The plus signal is ignored, the real function is: y = a + event_1×treat×beta1+ event_2×treat×beta2 + event_3×treat×beta3 + event_4×treat×beta4 + event_5×treat×beta5 + u y = a + event_1×treat×beta1 + u y = a + event_2×treat×beta2 + u y = a + event_3×treat×beta3 + u y = a + event_4×treat×beta4 + u y = a + event_5×treat×beta5 + u — Reply to this email directly, view it on GitHub <#25 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AE6EXFHWDG3HJ3RPA3Q2SO32ASVUNAVCNFSM6AAAAABRY7SIK6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZWGQ2TMMZSHA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>