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

st: RE: Interaction with an instrumented variable - IV


From   "Nichols, Austin" <ANichols@ui.urban.org>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Interaction with an instrumented variable - IV
Date   Thu, 26 Aug 2004 10:01:47 -0400

You've got two endogenous variables, just as if you had another endogenous
variable X4 and you did not realize that X4=X2*X3.  But if Z is a valid
instrument, and X3 is truly exogenous, then Z*X3 is a valid instrument, too.
You can use -ssc install ivreg2- and -ivreg2 `varlist', ffirst- to see if
your two instruments do in fact have enough power.  Also see Stock and Yogo
at http://www.nber.org/papers/T0284 for superior tests for the presence and
effect of weak instruments.  My guess is that you will wind up needing
another instrument W in the end.

-----Original Message-----
From: Ngo,PT (pgr) [mailto:P.T.Ngo@lse.ac.uk]
Sent: Thursday, August 26, 2004 9:29 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: Interaction with an instrumented variable - IV


Dear Statalisters,

I have a question re. IVs.
There was an earlier posting in the statalist archive, but it was not
conclusive. After some correspondence and a discussion today with a
colleague in my department, I am still unsure as to what is to be done. So I
am putting the question to you.

Y=b0+b1*X1+b2*X2+b3*X3+b4*X2*X3+U

X2 is the endogenous variable, which is instrumented with Z.
X2 is itself interacted with X3.
How does one model the interaction between an instrumented variable (X2
instrumented with Z) and an exogenous variable (X3)?

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index