Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Brian P. Poi" <brian@poiholdings.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: gmm with two equations and one parameter |
Date | Wed, 08 Jun 2011 07:57:28 -0400 |
On 06/07/2011 09:59 PM, Wu Zhang wrote:
OK. I attach all the files. my gmm command is: gmm (gmm_poi2) (gmm_poi3), mylhs(y) myrhs(x) instruments(1:d1, noconstant) instruments(2:d2, noconstant) vce(cluster id) myidvar(id) onestep
This represents a misunderstanding of the -gmm- syntax. You are combining elements of the interactive version of -gmm- along with elements of the function evaluator program version. If you write a program to evaluate your moment equations, then you write a single program that computes all of them; you cannot write separate programs for separate moment equations. Second, if you write a function evaluator program, then you do not bind its name in parentheses; the parentheses indicate to -gmm- that you are typing a moment equation interactively with substitutable expressions.
Here is a trivial example: ------- sysuse auto, clear program mygmm version 11 syntax varlist [if] , at(name) local eq1 : word 1 of `varlist' local eq2 : word 2 of `varlist' quietly replace `eq1' = mpg - `at'[1,1] - `at'[1,2]*turn quietly replace `eq2' = mpg - `at'[1,1] - `at'[1,3]*head end gmm mygmm, nequations(2) nparameters(3) /// instruments(turn head) winitial(identity) ------- -- Brian Poi -- brian@poiholdings.com * * 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/