Statalist The Stata Listserver


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

st: RE: Estimating fixed effects - why is a firm dropped?


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Estimating fixed effects - why is a firm dropped?
Date   Mon, 3 Apr 2006 17:49:39 +0100

Amy, 

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Amy Dunbar
> Sent: 03 April 2006 17:22
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Estimating fixed effects - why is a firm dropped?
> 
> I'm back with another panel data question. I am estimating 
> the following panel model.
> 
> btmit = a + ai + at + r0retit + r1retit + r2retit + r3retit + 
> r4retit + r5retit + r6retit  + eit
> 
> . tsset permno yeara
>        panel variable:  permno, 10051 to 92655
>        time variable:  yeara, 1993 to 1998
> 
> . xi: reg btm i.permno r0ret r1ret r2ret r3ret r4ret r5ret 
> r6ret i.yeara 
> i.permno        _Ipermno_10051-92655(naturally coded; _Ipermno_10051
> omitted)
> i.yeara         _Iyeara_1993-1998   (naturally coded; _Iyeara_1993
> omitted)
> 
> Another paper also uses this model.  The author states, "[The 
> firm fixed effect] is -ai in [the above equation], and 
> represents an intercept shift for each firm, incremental to 
> the overall cross-sectional intercept.  Thus, it is mean zero 
> by construction."
>
> My firm fixed effects do not sum to zero. Can someone help me 
> see what I am doing wrong? Thank you for your help.

You are using regress with explicit fixed effects.  My guess is that
regress is dropping one of the dummies, and the constant corresponds to
this category.  Hence the constant is not the "overall cross-sectional
intercept"; it's the intercept for the omitted category.  Try using
xtreg.  Also, it's not clear if you are using a balanced panel or not.
I think "mean zero by construction" applies strictly speaking to
balanced panels only, and you would need to use weights corresponding to
numbers of panel observations to get the fixed effects to sum to zero
otherwise.  The manual entry on xtreg, if I'm not mistaken, has a
discussion of this.

HTH.

--Mark

> 
> Amy Dunbar
> University of Connecticut
> 
> *
> *   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/
> 
> 

*
*   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