Statalist The Stata Listserver


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

Re: st: between overall within squared on xtreg


From   Justin Smith <[email protected]>
To   [email protected]
Subject   Re: st: between overall within squared on xtreg
Date   Tue, 15 Aug 2006 22:34:57 -0400

The three R-squareds are:

1) within:  The R-squared from the mean-deviated regression, i.e. the
ordinary r-squared from running OLS on the transformed data.

2)between: first, this computes the fitted values using the
fixed-effects parameter vector and the within-individual means of the
indepdenent variables.  Then calculates the r-squared as the squared
correlation between those predicted values and the within-individual
means of the original y variable.

3) overall: first, this computes the fitted values using the
fixed-effects parameter vector and the original, untransformed
independent variables.  Then calculates the r-squared as the squared
correlation between those predicted values and the original,
untransformed y variable.


Seems to me like you would want the within R-squared, since you are
using the within estimator.

Justin Smith




On Wed, 16 Aug 2006 01:54:44 +0200
 "Steffen H.  Gr�ning" <[email protected]> wrote:
JS>Hi,
JS>
JS>I have been looking on FAQ and quite every other resource that
JS>google
JS>pointed me to. I have gotten so far as to know that when using a
JS>fixed
JS>effects model to only interpret the within r squared.
JS>But why is stata giving the other r squares as well and how can i
JS>interpret them?
JS>I haven�t found a clear cut answer to the differences and
JS>interpretations of within between and overall r squared. Maybe
JS>someone
JS>could help me out here?
JS>
JS>Thank You.
JS>Steffen.
JS>
JS>
JS>*
JS>*   For searches and help try:
JS>*   http://www.stata.com/support/faqs/res/findit.html
JS>*   http://www.stata.com/support/statalist/faq
JS>*   http://www.ats.ucla.edu/stat/stata/

Justin Smith
PhD Candidate
Department of Economics
McMaster University
Phone: (905) 962-0353
E-mail: [email protected] 
*
*   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