Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: RE: FE OLS with few clusters
From
"Schaffer, Mark E" <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: RE: FE OLS with few clusters
Date
Wed, 5 Jun 2013 10:20:28 +0000
Marc,
You can run an FE estimation, but probably want to use -areg-. It will give you the same coefficients as -xtreg,fe- but the setup assumes a fixed number of panels (countries in your case) and with the number of observations increasing with the sample size. From -areg- help:
areg fits a linear regression absorbing one categorical factor. areg is designed for datasets
with many groups, but not a number of groups that increases with the sample size.
BTW, -areg- doesn't require "many groups", so you are OK there.
But in fact, with only 8 panels, you could simply include country dummies by hand and get the same results. -areg- will report a mean intercept for the sample as a whole, but otherwise there's no particular advantage to using it when you have so few panels.
HTH,
Mark
> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of Marc Peters
> Sent: 29 May 2013 21:50
> To: [email protected]
> Subject: st: FE OLS with few clusters
>
> Dear all,
>
>
> I have a dataset with units clustered into countries and want to specify a
> fixed effect regression model. My problem is that the number of countries in
> my dataset is very small (only 8) while the number of observations/country is
> larger (n=40-350), do I have too few clusters to run a FE OLS regression? If so,
> is there some other kind of model I could run?
>
> Thanks for your help,
>
>
> Marc
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/statalist-faq/
> * http://www.ats.ucla.edu/stat/stata/
-----
Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)
We invite research leaders and ambitious early career researchers to
join us in leading and driving research in key inter-disciplinary themes.
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/