Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: (too?) Low within R-sq in FE and clustered errors.
From
"Anat (Manes) Tchetchik" <[email protected]>
To
[email protected]
Subject
st: (too?) Low within R-sq in FE and clustered errors.
Date
Sun, 3 Nov 2013 01:06:34 +0200
Dear Statalist members,
I ran a FE regression whose relevant to my question results are: .
Fixed-effects (within) regression Number of obs = 438
Group variable: country_code Number of groups = 35
R-sq: within = 0.0796 Obs per group: min = 6
between = 0.4188 avg = 12.5
overall = 0.4087 max = 16
F(6,34) = 5.31
corr(u_i, Xb) = 0.0601 Prob > F = 0.0006
I have 2 quick questions:
(1) I'm a bit concerned that the "within R-sq" is pretty low.
Does it mean that I gain no benefit from using a panel structure?
(2) I used clustered errors, but having read "Clustered Errors in Stata" (
A. Nichols and M. Schaffer) I'm not sure whether my 35 groups and
pretty unbalanced panel, justify using such errors.
Your responses will be greatly appreciated!
Best,
Anat Tchetchik, PhD
Department of Business Administration
Guilford Glazer Faculty of Business and Management
Ben-Gurion University of the Negev
P.O.Box: 653
Beer-Sheva, Israel, 84105
E-mail: [email protected]
Phone 972-(0)8-6479735
Fax: 972-(0)8-6472920
--
Anat Tchetchik, PhD
Department of Business Administration
Guilford Glazer Faculty of Business and Management
Ben-Gurion University of the Negev
P.O.Box: 653
Beer-Sheva, Israel, 84105
E-mail: [email protected]
Phone 972-(0)8-6479735
Fax: 972-(0)8-6472920
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/