Statalist


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

st: sargan test and hansen j statistic- xtabond2


From   Allison.Milner@student.griffith.edu.au
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: sargan test and hansen j statistic- xtabond2
Date   Fri, 20 Feb 2009 08:41:40 +1000

Hi,

I am trying to run xtabond2 (http://fmwww.bc.edu/RePEc/bocode/x/xtabond2.hlp) on an unbalanced panel that has 74 countries from 1980 to 2006.
Regardless of how I alter the lag structure of the my lagged DV, I obtain a non-significant hansen J statistic and a significant sargan statistic. The hansen test performs best if I limit the number of instruments that I am using. I know that these tests are crucial in the argument that the instruments being used are exogenous but do I need to find consistent results using both tests?

here is an example of the commands I am using:
xtabond2 sq_FemalesASR iGini ilog_healthpercap ilog_unieduexp ilog_fert ilog_eduexpGDPpercap year totalG, r two gmm(sq_FemalesASR, lag(1 2)) iv(iGini year ilog_healthpercap ilog_unieduexp ilog_fert ilog_eduexp GDPpercap  year totalG)

** I have also tried this using 'collapse' after the gmm variable.

Thanks for the help in advance,

Allison


*
*   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   |   What's new   |   Site index