RE: Antwort: Re: st: RE: Use fixed-effects' coefficients in a second regression

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Actually, you might not want to go down that route - the P-T estimator
has been heavily criticized by some heavyweights in the field.  See
Kit's post from a couple of weeks ago:
 
http://www.stata.com/statalist/archive/2011-01/msg00220.html

--Mark

________________________________

	From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Justina
Fischer
	Sent: 20 January 2011 22:24
	To: statalist@hsphsun2.harvard.edu
	Subject: Antwort: Re: st: RE: Use fixed-effects' coefficients in
a second regression
	
	

	Hi Fabien,
	
	if not already suggested, you miht consider the estimator
developed by Pluemper and Troeger.
	
	For more info, see here:
	
	http://www.polsci.org/pluemper/ssc.html
	
	Best
	Justina
	
	-----owner-statalist@hsphsun2.harvard.edu schrieb: -----
	
	An: statalist@hsphsun2.harvard.edu
	Von: Austin Nichols <austinnichols@gmail.com>
	Gesendet von: owner-statalist@hsphsun2.harvard.edu
	Datum: 20.01.2011 06:40PM
	Thema: Re: st: RE: Use fixed-effects' coefficients in a second
regression
	
	Fabien Bertho:
	Also note that the estimated FE will be measured with error and
you
	may want to weight your second regression appropriately using an
	estimate of variance or "shrink" the FE toward some group mean;
see
	also
	findit fese
	findit felsdvreg
	findit felsdvregdm
	
	On Thu, Jan 20, 2011 at 11:32 AM, Schaffer, Mark E
	<M.E.Schaffer@hw.ac.uk> wrote:
	> Fabien,
	>
	>> -----Original Message-----
	>> From: owner-statalist@hsphsun2.harvard.edu
	>> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of
	>> Fabien Bertho
	>> Sent: 20 January 2011 14:23
	>> To: statalist@hsphsun2.harvard.edu
	>> Subject: st: Use fixed-effects' coefficients in a second
regression
	>>
	>> Hi,
	>>
	>> I would like to run an estimation in two steps. The first
	>> step consists in running a regression with control variables
	>> and country fixed effects. Then, I would like to convert
	>> fixed-effects&#8217; coefficients in variables in order to
	>> use them as dependent variables in a second regression.
	>
	> You can get them with -predict- and the -u- option.
	*
	*   For searches and help try:
	*   http://www.stata.com/help.cgi?search
	*   http://www.stata.com/support/statalist/faq
	*   http://www.ats.ucla.edu/stat/stata/
	
	



-- 
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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/