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

st: Fwd: -ivreg2-how to instrument multiple endogenous variables

From   Nirina F <>
Subject   st: Fwd: -ivreg2-how to instrument multiple endogenous variables
Date   Tue, 2 Jun 2009 14:35:20 -0400

Sorry, it got sent however I was not finished explaining my questions:

I run in stata ivreg2 to instrument four mutually exlcusive
indicators. I had four instrumental variables so I got some results
but I have no idea how to interpret the underidentification and weak
instrument test. When I increase the number of instruments, I get the
hansen's J but agian, what does it mean now?
Does anyone know how I should deal with multiple endogenous variables?

A second question, the coefficients  I get from OLS on these four
variables are proportionately higher than in 2SLS, meaning
var1 0.2
var2 0.3
var3 0.5
var4 0.5
in 2SLS
I got
var1 0.4
var2 0.4
var3 0.6
var5 0.7
So should I suspect some problem in the 2SLS given that for exemple
the difference btween var1 and var3 is higher in OLS than in 2SLS?
Thank you in advance,

*   For searches and help try:

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