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

st: RE: xtabond2: Proper use of ivstyle(.)


From   "Salvati, Jean" <JSalvati@imf.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: xtabond2: Proper use of ivstyle(.)
Date   Tue, 17 Jan 2006 09:51:39 -0500

Niko,

I believe the issue is that the instruments matrix is not constructed
exactly the same way in both cases. 

With < iv(l2.X1, equ(level)) iv(l2d.X1, equ(diff)) >, the instruments
for the levels equation and for the FD equation are in separate columns.
With < iv(l2.X1, equ(both)) > they're in the same column.

Does the instruments matrix have one more column in the first case than
in the second case?

Jean


> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Niko Wrede
> Sent: Monday, January 16, 2006 9:36 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: xtabond2: Proper use of ivstyle(.)
> 
> I am using xtabond2 and I want to use the ivstyle supoptions 
> equ(level) and
> equ(diff) in the proper way.
> Therefore I thought that following commands should produce 
> the same results:
> 
> xtabond2 Depvar l2.Depvar l2.X1, gmm(Depvar, lag(3 .) 
> equ(both)) iv(l2.X1,
> equ(both)) twostep arte(3) small rob
> 
> IS EQUAL TO
> 
> xtabond2 Depvar l2.Depvar l2.X1, gmm(Depvar, lag(3 .) 
> equ(both)) iv(l2.X1,
> equ(level)) iv(l2d.X1, equ(diff)) twostep arte(3) small rob
> 
> 
> OR PUT IN ANOTHER WAY
> 
> ... iv(l2.X1, equ(both)) ... IS THE SAME AS ... iv(l2.X1, 
> equ(level)) iv(l2d.X1, equ(diff)) ...
> 
> Reason: Help file of xtabond2 says for the equ(both) command 
> that "...  Also by default, the instruments are 
> first-differenced for use in the
>         first-difference equations and taken as is for 
> instrumenting the levels equations. "
> 
> However, results differ. Does anybody know, when the extended 
> version with iv( ., equ(level)) and iv(., equ(diff)) 
> generates the same results as the iv(., equ(both)) command?
> 
> Many thanks in advance
> Niko
> 
> 
> *
> *   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/
> 

*
*   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 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index