Page 1 of 1
Two way fixed effects and unbalanced panel
Posted: Thu Apr 09, 2015 2:42 am
by jocob
Hi there:),
I'm new here and not really an Eviews pro, but I hope you guys can help me with an issue.
I have an unbalanced panel (for 75 companies and 15 years) and I need to do a cross-section fixed and period fixed effects.
But I get the message that it is not possible for an unbalanced panel. Anyway to solve this issue?
Thanks in advance!
Re: Two way fixed effects and unbalanced panel
Posted: Thu Apr 09, 2015 4:02 am
by EViews Gareth
Can you provide your workfile?
Re: Two way fixed effects and unbalanced panel
Posted: Thu Apr 09, 2015 4:32 am
by jocob
Is that necessary? It's a bit confidential.
Re: Two way fixed effects and unbalanced panel
Posted: Thu Apr 09, 2015 9:41 am
by EViews Glenn
Unbalanced two-way fixed effects should be supported. Are you doing any weighting or a special coefficient covariance calculation?
Re: Two way fixed effects and unbalanced panel
Posted: Sat Apr 11, 2015 6:42 am
by jocob
No nothing strange,
But I'm sorry I don't know what you precisely mean with weighting or special calculations?!
Re: Two way fixed effects and unbalanced panel
Posted: Sun Apr 12, 2015 8:32 am
by EViews Glenn
Robust covariances or some form of SUR-type estimation.
Re: Two way fixed effects and unbalanced panel
Posted: Mon Apr 13, 2015 5:24 am
by jocob
No haven't done that.
So I don't know what the problem is.
But it should be possible with an unbalanced panel to do cross-section and period fixed effects at the same time? Because the error says it's not possible.
Re: Two way fixed effects and unbalanced panel
Posted: Mon Apr 13, 2015 6:59 am
by EViews Gareth
Yep, perfectly possible.