Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: p-values based on non-standard critical t-values after -xtreg, fe- and using -estout-


From   Bert Jung <bjung59@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: p-values based on non-standard critical t-values after -xtreg, fe- and using -estout-
Date   Fri, 7 Dec 2012 16:10:46 -0500

Dear Statalisters,

I run -xtreg, fe- regressions with a small number of clusters and
would like to adjust the critical values along the lines of Donald and
Lang (2007), using a t(G-2) distribution where G is the number of
clusters.  In my case the number of clusters is constant across all
regressions at G=20.

According to Baum et al (2011) Stata uses a t(G-1) distribution to
determine critical t-values and hence the p-values.  I am looking for
an efficient and transparent way to instead output p-values using
t(G-2).

I use Ben Jann's user-written -esttab- (on SSRN).

Any thoughts much welcome.
Thanks Bert


Donald SG, Lang K. Inference with Difference-in-Differences and Other
Panel Data. Review of Economics and Statistics 2007; 89; 221–233.

Baum et al 2011, "Evaluating one-way and two-way cluster-robust
covariance matrix estimates"
http://fmwww.bc.edu/repec/dsug2011/desug11_schaffer.pdf

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index