Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Gerek, Caner" <cg205@leicester.ac.uk> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st:overidentification - system gmm |
Date | Wed, 7 Aug 2013 01:37:31 +0100 |
dear statalist; I am using the xtdpdsys stata command below for system gmm estimation, xtdpdsys netintincomeASSETS inflation gdpgrowth Lerner EquityASSETS lnloans feesASSETS trade cashASSETS efficiency, lags(1) maxldep(1) twostep pre(Lerner) endog(feesASSETS) artests(2) but I have a overidentification problem even my # of instruments is 83 and # of observation is 2657. I tried to convert it to xtabond2 as xtabond2 netintincomeASSETS L.netintincomeASSETS inflation gdpgrowth Lerner EquityASSETS lnloans feesASSETS trade cashASSETS efficiency, gmm(L.netintincomeASSETS Lerner L.feesASSETS, lag (0 0)eq(level)) gmm(L.netintincomeASSETS Lerner L.feesASSETS, lag(1 1) equation(diff)) iv( inflation gdpgrowth EquityASSETS lnloans trade cashASSETS efficiency) robust twostep nocons artests(2) what is my mistake? I converted it to xtabond2 to reduce instruments by collapse option but I put it to each gmm() and system says invalid. what should I do to remove this overidentification problem thanks in advance Caner Gerek University of Leicester * * 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/