Rafa,
From: "R.E. De Hoyos" <[email protected]>
To: <[email protected]>
Subject: st: Fixed effects and e(sample)
Date sent: Wed, 19 May 2004 16:13:31 +0100
Send reply to: [email protected]
> We are using panel data and for some reason when I run fixed effects or
> random effects the sample size is different. My hypothesis here is that
> observations with one observation are dropped automatically but I haven't
> thought about this enough.
I tried to replicate the phenomenon by creating a dataset in which
one group has one observation, and failed. Both re and fe used the
same number of observations.
Can you post the output of the two commands, if it isn't too bulky?
(If the output is voluminous, you can send it to me privately.) Also
check which version of xtreg you're using (-which xtreg-).
--Mark
> However stata seems to only tell me it's dropping observations in the
> observation count when I run fixed effects but is actually keeping the
> same number of observations in e(sample) as the random effects has.
>
> Is there a problem in using e(sample) when using fixed effects? If so what
> variable stores the values?
>
> Thanks,
>
> Rafa
> *
> * 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/
Prof. Mark E. Schaffer
Director
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
http://www.som.hw.ac.uk/cert
*
* 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/