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

st: RE: xtivreg instruments

From   "Steven Stillman (LMPG)" <>
To   "''" <>
Subject   st: RE: xtivreg instruments
Date   Wed, 30 Jul 2003 12:04:24 +1200

This questions seems to come up quite often.  In fact, there is a related
FAQ at  The bottom line
is that all excluded exogenous variables must be used as instruments for all
endogenous variables in any IV model.  In addition, even if there is a
structural relationship, one endogenous variable can never be used as an
instrument for another.  This can be clearly seen if you take your
underlying structural model and rewrite it as a reduced-form model with only
exogenous variables as independent variables.   The only exception to this
is when you have a recursive or triangular system (the FAQ explains this in
more detail), but this does not sound like your situation.


> -----Original Message-----
> From:	Felipe Targa []
> Sent:	Wednesday, July 30, 2003 2:58 AM
> To:
> Subject:	st: xtivreg instruments
> I am estimating a GLS random-effects for panel data using the xtivreg
> STATA command. I have two endogenous variables. However, in my
> specification these two endogenous variables are affected by a different
> set of exogenous variables.  STATA takes the exogenous variables affecting
> the first endogenous variable as regressors for the second endogenous
> variable. I would like to know what I can do to restrict the estimation of
> the first endogenous variable with ONLY the exogenous variables specified
> in the command (one of them is the other endogenous variable).
> Thank you,
> Felipe Targa
> *
> *   For searches and help try:
> *
> *
> *

The information contained in this document is intended only for the
 addressee and is not necessarily the views nor the official 
communication of the Department of Labour.  All final/official papers 
which are sent from the Department will be sent by non-electronic
means, on appropriate letterhead, signed by authorised personnel.
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index