Statalist The Stata Listserver


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

st: Re: xtrc -- Should the (mean) coefficients be different than those ofxtreg, fe, and if so, why and in what direction?


From   "Michael Blasnik" <[email protected]>
To   <[email protected]>
Subject   st: Re: xtrc -- Should the (mean) coefficients be different than those ofxtreg, fe, and if so, why and in what direction?
Date   Fri, 01 Jun 2007 13:12:52 -0400

...
Wow - brief Statalist post ever? (just a subject)
1) Yes, xtrc coefficients can be expected to differ from xtreg, fe
2) Why? - because it's a different estimator? Read the manual (or any econometrics text).
3) Direction - it depends...

A little less briefly - the fe model is basically pooled OLS after de-meaning within each panel. The xtrc model is more like a weighted average of the coefficient estimates from individual regressions run within each panel separately. Panels with the better-determined coefficient estimates get greater weight in determining the overall (mean) coefficient estimate. So if the coefficient for a certain variable tends to vary with the amount of "noise" in the panel, then the difference can be systematic.

Michael Blasnik

----- Original Message ----- From: "david reinstein" <[email protected]>
To: "statalist" <[email protected]>
Sent: Friday, June 01, 2007 12:22 PM
Subject: st: xtrc -- Should the (mean) coefficients be different than those of xtreg, fe, and if so, why and in what direction?


*
* 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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index