ed?
Date: Tue, 3 Oct 2006 09:29:01 -0700
Message-ID: <[email protected]=
u>
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
Thread-topic: hierarchical F-tests and fixed effects regression -- which R-=
Squared?
Thread-index: AcbnCFJfXMLNqvofSVuCvudNKAU+Bg=3D=3D
From: "Jason Yackee" <[email protected]>
To: <[email protected]>
X-Barracuda-Virus-Scanned: by Barracuda Spam Firewall at usc.edu
X-Barracuda-Spam-Score: 0.00
X-Barracuda-Spam-Status: No, SCORE=3D0.00 using global scores of TAG_LEVEL=
=3D5.0 QUARANTINE_LEVEL=3D5.4 KILL_LEVEL=3D8.0
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by hsphsun2.harvard.edu=
id k93GRF822169
Sender: [email protected]
Precedence: bulk
Reply-To: [email protected]
Errors-To: [email protected]
Dear all,
Apologies if this is a double post; my university's servers have a dislike =
for Statalist that I am trying to sort out.
I would like to manually calculate a hierarchical F-test to see if adding a=
n interaction term to my equation significantly supplements the explained v=
ariance.=A0 I am using -xtgls- with the fixed effects option.=A0 I am aware=
that the "reported" r-squared statistic in publications is usually the "wi=
thin" statistic. =A0What I am wondering is which is the proper r-squared to=
plug into the standard f-statistic formula,
F =3D [(RSquare2 - RSquare1)/k2-k1)] / [(1-RSquare2)/(N-k2-1).
I am guessing that I should put in the "overall" r-squared from the -xtgls-=
output, but that is really just a guess.... Advice appreciated as usual
Jason Webb Yackee
Fellow
Gould School of Law
University of Southern California
Los Angeles, California 90089-0071
Tel: 213-740-2865
Fax: 213-740-5502
Cell: 919-358-3040
email: [email protected]
=A0
Jason Webb Yackee, PhD Candidate; J.D.
Fellow, Gould School of Law
University of Southern California
[email protected]
Cell: 919-358-3040
*
* 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/
