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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: negative R-squared in gmm


From   Christopher Baum <kit.baum@bc.edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: negative R-squared in gmm
Date   Sun, 25 Jul 2010 06:25:35 -0400

<>
On Jul 25, 2010, at 2:33 AM, Jing wrote:

> I am using "xtivreg2 dep indep (indep=iv) , fe gmm bw (1) robust" to my data. but the centered and uncertered R-squared are both negative. Could you please advise me whether this is acceptable. if not, what are the potential problems, and how can I refine my model?

Negative r^2 measures in any instrumental variables estimation are quite common. IV does not minimize the sum of squared residuals (y- Xb ), but rather the sum of squared residuals (y - Zb). Using GMM, the same issue arises. And in a fixed effect model, there is no single definition of r^2 anyway.

By the way bw(1) in this context does absolutely nothing. If you want to allow for one period of serial dependence, you want bw(2). 

Kit



Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index