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]

st: AW: new features for cmp and xtabond2


From   "Dithmer, Jan" <jdithme@food-econ.uni-kiel.de>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: AW: new features for cmp and xtabond2
Date   Wed, 18 Apr 2012 13:14:44 +0200

Dear David,

I really like the idea to use principal components analysis to reduce the instrument count.
However, when the panel under scrutiny has many time periods, the instrument set may still be large.
Therefore, may the 'pca' option be combined with the 'collapse' or the 'laglimits' option to further reduce the number of instruments?

Best, Jan


-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von David Roodman (droodman@cgdev.org)
Gesendet: Tuesday, April 17, 2012 3:12 AM
An: statalist@hsphsun2.harvard.edu
Betreff: st: new features for cmp and xtabond2

I've added the following features:
* cmp now has a covariance option that makes it easier to impose certain constraints on the error covariance matrix for each level of a model. So in a two-level model "covariance(independent unstructured)" would specify that the random effects are not correlated across equations while leaving the covariances at the observation level unconstrainted. As another example, "covariance(exchangeable exchangeable)" specifies that at each level all correlations between pairs of errors are the same. See the help for xtmixed for more on these ideas.
* xtabond2 now accepts factor variables.
* xtabond2 now offers two options to use principal components analysis to reduce the instrument count. Option "pca" specifies that xtabond2 replace the "GMM-style" instruments with their principal components. By default, it keeps all components with eigenvalue greater than 1, or more if necessary to prevent underidentification. Option "components()" allows you to override this default. The help file contains an example.

Install each by typing "ssc install [program name], replace" and restarting Stata.
--David


David Roodman
Senior Fellow
1800 Massachusetts Ave. NW
3rd floor
Washington, DC 20036
Center for Global Development
Independent Research and Practical Ideas for Global Prosperity
 
Phone: 1 202 416-4023
http://blogs.cgdev.org/open_book



*
*   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/

*
*   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