How to optimize fixed effects model (Diff in Diff)?

[Daniel Ciobanu]

Hi everyone!

Is there anyone who can help me optimize a STATA code? I am running a region specific time varying fixed effects model which takes ages using 'reg'. Is there a way to run this more efficiently?

I use a Diff in Diff technique. This is my regression:

reg outcome treated##post1 i.region##i.year, vce(cl region)

My panel data includes 391 regions and covers a period of 20 years. Therefore, the i.region#i.year yields a big number of combinations. This makes the regression very time consuming.

[Jeff Wooldridge]

Is the treatment at the regional level? If so, this regression doesn’t identify a treatment effect.

Maybe you have units within region, some treated, some not. Is that true?

[Nick Cox]

Cross-posted at https://www.reddit.com/r/stata/comme...fixed_effects/

It's a rule at Reddit and a request here that you tell people about cross-posting elsewhere.

[Daniel Ciobanu]

It is true Jeff, the treated units are grid cells within regions!

[Daniel Ciobanu]

SOLVED: For those interested, "reghdfe" seems to do the trick!