S.E. in xtreg and areg

[Cooper Felix]

Dear friends,

I'm estimating a fixed effects model with an unbalanced panel data using both areg and xtreg commands, the estimated coefficients are the same as expected but standard errors are different making one of the key coefficients highly significant in areg but insignificant in xtreg. Generally speaking, which command is more reliable and which one should I use?

Thanks!

[Joro Kolev]

I think this should not be happening.

Copy and paste here the exact commands you executed, and the exact output Stata returned to you .

[FernandoRios]

Dear Cooper
If you are using the default standard errors, both areg and xtreg will provide the same results. When using robust or cluster standard errors, results will be different because XTREG uses a Sandwich formula that "produces a consistent VCE estimator when the disturbances are not identically distributed over the panels or there is serial correlation in eit".
Areg, however, does not account for any serial correlation.
If your data is indeed panel data, xtreg should be the right way to go.
HTH
Fernando

[Cooper Felix]

Dear Cooper
If you are using the default standard errors, both areg and xtreg will provide the same results. When using robust or cluster standard errors, results will be different because XTREG uses a Sandwich formula that "produces a consistent VCE estimator when the disturbances are not identically distributed over the panels or there is serial correlation in eit".
Areg, however, does not account for any serial correlation.
If your data is indeed panel data, xtreg should be the right way to go.
HTH
Fernando

Thanks for your input, yes, I did use robust SE in my estimation.