reghdfe and xtreg commands yield different coefficients

[TramAnh Nguyen]

Hello Statalist users,

Let me take an example of the problem I encounter. I have a panel dataset and I want to regress profit on firm size and firm age with year fixed effects model. I use three commands as follows:

xtset firmid year
reghdfe profit size age, absorb(year) cluster(firmid)
xtreg profit size firmage i.year,  vce(cluster firmid)
reg profit size age i.year,  vce(cluster firmid)

reghdfe and reg give me similar coefficients, which is different from the result of xtreg. So can reghdfe replace xtreg in my case?

Thank you in advance.
Tram-Anh Nguyen.

[Carlo Lazzaro]

Tram-Anh:
short answer: because you ran three different regression codes (-fe- ; -re-; pooled OLS).
More substantively: why not posting on the General forum, as per FAQ? Thanks..

[TramAnh Nguyen]

Dear Dr. Carlo Lazzaro,
Thank you very much for the reply. And I am so sorry because I have just realized that I clicked on the wrong tab- Sandbox instead of General tab to post my question.

Kind regards,
Tram-Anh Nguyen.