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 at the end of May, and its replacement, statalist.org is already up and running.


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

st: xtabond2 with just-identified model


From   "Lovisa Persson" <lovisa.persson@nek.uu.se>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: xtabond2 with just-identified model
Date   Thu, 20 Sep 2012 13:05:20 +0200

Hello users,

What does it mean to run a just-identified model with gmmstyle instruments
using xtabond2?
How come it produces different results from when you run ivstyle
regressions? 

I run my just-identified model like this:

xtabond2 kostnader intäkter dum1-dum17 lnpopulation lnzerofour
lnsixtyfiveplus, gmm(intäkter, laglimits(3 3)) iv(dum1-dum17 lnpopulation
lnzerofour lnsixtyfiveplus) robust noleveleq artests(3)

How come the overidentification tests are run even though I only use one
instrument? And why are the results not the same as the following two codes
using ivreg2 and xtabond2:

ivreg2 D.kostnader dum1-dum16 D.(lnpopulation lnzerofour lnsixtyfiveplus)
(D.intäkter=L3.intäkter), cluster(kod)

xtabond2 kostnader intäkter dum1-dum16 lnpopulation lnzerofour
lnsixtyfiveplus, iv(dum1-dum16 lnpopulation lnzerofour lnsixtyfiveplus)
iv(L3.intäkter, passthru) robust noleveleq artests(3)

The two codes above produce the same results?

There is some intuition here that I am not getting about the gmmstyle
instrument matrix. 
Is there someone out there that can help me sort out the concepts here?

Thank you!

Lovisa Persson
Ph.D. Student 
Department of Economics 
Uppsala University 
E-mail: Lovisa.Persson@nek.uu.se



*
*   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