Statalist The Stata Listserver


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

st: Re: xtreg areg and R-square values


From   "Rodrigo A. Alfaro" <ralfaro76@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: xtreg areg and R-square values
Date   Tue, 11 Apr 2006 14:56:33 -0400

1. The R-squares between xtreg (fe) and areg are differt. I looked at the explanations on the stata faq. It seems to suggest that the xtreg command does not include the explanatory power of the dummy variable. However areg in my cases seems to provide a considerably higher R-square as compared to xtreg. my Areg r-squares are of the order of .25 and xtreg are .09. (overall) Does that mean that my indepent variables do not really contribute to explaining a lot of variation ?. What is the correct thing to report.
Rod: areg uses dummy variables to compute R-2, xtreg shows 3 different R-2... there is not a unique measure, but usually within R-2 is reported.


I did see a query like this in stata listserv earlier. I dont think it provided a complete answer.
Rod: full description of the R-2 for xtreg is available in the reference manual. in the case of areg the R-2 is the same as reg.


2. the xtreg fe command in stata provides values of rho and values of standard deviation of the the random intercept. Why is that?. My basic knowledge is that fixed effect models dont have such values. For example areg does not provide these. What is the utility of these in the context of fixed effect panel data regression ?
Rod: in the reference manual (command xtreg) is explained this topic, after example 2.
*
* 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