Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: RE : st: Fixed-effects, unbalanced panel and time-invariant va=


From   Vera Troeger <vtroe@essex.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: RE : st: Fixed-effects, unbalanced panel and time-invariant va=
Date   Tue, 17 Oct 2006 16:14:14 +0100

riable
References: <E12BB747ABB333469FA7F762DDA14AB18A63F9@BOLOGNE.ccinca.fr>
In-Reply-To: <E12BB747ABB333469FA7F762DDA14AB18A63F9@BOLOGNE.ccinca.fr>
Content-Type: text/plain; charset=3DISO-8859-1; format=3Dflowed
Content-Transfer-Encoding: 8bit
X-Essex-ClamAV: No malware found
X-Essex-MailScanner: Found to be clean
X-Essex-MailScanner-SpamCheck: not spam, SpamAssassin (score=3D-1.113,
  =09required 5, ALL_TRUSTED -1.44, TW_RX 0.08, VBOUNCE2_SUBJ_AV 0.25)
X-Essex-MailScanner-From: vtroe@essex.ac.uk
Sender: owner-statalist@hsphsun2.harvard.edu
Precedence: bulk
Reply-To: statalist@hsphsun2.harvard.edu
Errors-To: statalist-owner@hsphsun2.harvard.edu
yes, it is an ongoing discussion. however, the SEs of the latest version
should be correct, at least we tested this by MC simulations. the SEs
for time varying variables are calculated with a demeaned VC matrix and
the degrees of freedom are corrected for regressors plus n-1 unit
effects. the SEs for time invariant variables are pooled ols with
corrected degrees of freedom for no of regressors plus n-1 unit effects.
it is described in the paper. with simulated data the SEs were correct
for various settings and specifications.
Vera

PETITT Barbara schrieb:
> Thanks, Vera. Did you get my other message about the computation of the s=
tandard errors for xtfevd? Somehow, it looks like the coefficient are fine,=
  but the standard errors and the t-stats are weird. Thanks again. Barbara.
>
> -----Message d'origine-----
> De : owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun=
2.harvard.edu] De la part de Vera Troeger
> Envoy=E9 : mardi 17 octobre 2006 16:12
> =C0 : statalist@hsphsun2.harvard.edu
> Objet : Re: st: Fixed-effects, unbalanced panel and time-invariant variab=
le
>
>
> xtfevd does work for unbalanced panels in the same way as fe. Vera
>
> PETITT Barbara schrieb:
>
>> Hello,
>>
>> I have an unbalanced panel (for some firms, I have three years of
>> data, for others, two years or even only one year). One of my
>> independent variables is a time-invariant variable and with
>> fixed-effects models, its gets dropped. I contemplated using the fixed
>> effects vectorxtfevd decomposition (xtfevd), but does it work for
>> unbalanced panel? Otherwise, is there any alternative? Thank you.
>>
>> Barbara.
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/support/faqs/res/findit.html
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>>
>>
>>
>>
>
>

--=20
Vera E. Troeger
Lecturer in Political Science
Government Department
University of Essex
Wivenhoe Park
Colchester CO4 3SQ
UK

phone: +44 (0)1206 872509
email: vtroe@essex.ac.uk


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index