Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: re: xtabond2


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: re: xtabond2
Date   Fri, 1 Apr 2005 21:15:31 -0500

Dahlia said

I am new to Statalist. I am working on my PhD dissertation in
Economics at George Washington University. I would like to use
xtabond2 to estimate an Euler Investment Equation using the GMM system
estimator.

I leaned about xtabond2 from the internet while looking for some
Econometric papers (as it is not part of STATA 8.0 which I am
currently using).

My dissertation advisor is asking me if this command "xtabond2" has
been validated or authorized ?.


In private correspondence, I suggested that Dahlia contact xtabond2 author David Roodman (who is also located in DC), and that the xtabond2 help file contains several examples that allow you to generate the same results (using the same data) as does Doornik's DPD for Ox. In that sense I consider that the xtabond2 routine is validated, in that the author has provided benchmarks for which his routine matches DPD for Ox (except for one issue, where Roodman believes that Doornik;s code is off by a factor of two on a particular stat: still an unsolved mystery). The validation of xtabond2 against alternative software exceeds the level of validation commonly provided for user-contributed Stata code IMHO.

As for the notion of "authorized", it is not relevant for user-contributed code in the Stata user community. No one is able to authorize or deauthorize user-written routines. As I have stated before on this list, if users raise issues with a particular routine, and the author(s) of that routine do not respond to enquires or bug reports (or are no longer involved with developing the code), then I will withdraw the routine from the SSC archive. That does not prevent software from being distributed, as any user may make Stata routines available from their own pages. But inclusion of a routine in SSC does not provide any assurance that it does what it is supposed to do. That is the author(s)' assertion, not mine, and users of the software do so at their own risk. But then that is true even for commercial software, is it not? A couple of long-lived bugs in Excel functions come to mind...

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html

*
* 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/




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