Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: hausman test after reg3


From   Val�rie Orozco <[email protected]>
To   [email protected]
Subject   st: hausman test after reg3
Date   Tue, 09 Jan 2007 15:14:44 +0100

Hi,

I have a question about doing the test of endogeneity (hausman test) after a "reg3" estimation.

I have a first estimation of a system of 3 equations where I add 5 equations of instrumentation. It looks like this :

reg3 (dw1 dY dp1 dp2 dp3 dp4 t1 t2)/*
*/(dw2 dY dp1 dp2 dp3 dp4 t1 t2)/*
*/(dw3 dY dp1 dp2 dp3 dp4 t1 t2)/*
*/(Y Ylag pib t1)/*
*/(p1 iv1 iv2 iv3 iv4 iv5)/*
*/(p2 iv1 iv2 iv3 iv6 iv7)/*
*/(p3 iv8 iv9 iv1 iv2 iv3 iv10 iv11 iv12 iv13)/*
*/(p4 iv1 iv2 iv3 iv14 iv15)/*
*/ , cons(1-16) /*I have some constraints*/

est store reg_with_iv

dw1, dw2 and dw3 are my dependant variables
dY, dp1, dp2, dp3 and dp4 are supposed to be endogeneous (Y, p1, p2, p3, p4 are the same variables not transformed in differences)
Ylag, pib and iv`j' with j=1,...,15 are a set of instruments

To test the endogeneity in my system, I run the same system without the five equations of instrumentation and run the hausman test doing :

reg3 (dw1 dY dp1 dp2 dp3 dp4 t1 t2)/*
*/(dw2 dY dp1 dp2 dp3 dp4 t1 t2)/*
*/(dw3 dY dp1 dp2 dp3 dp4 t1 t2)/*
*/ , cons(1-16)
est store reg_without_iv

hausman store reg_with_iv reg_without_iv , alleqs constant

But the problem is that since I haven't the same number of parameters estimated, the variance covariance matrix haven't the same dimension, and the test failed...
Is there a better and right way to deal with endogeneity after "reg3" ????

Thank you for trying to help me.

val�rie


--

***********************
Val�rie Orozco
INRA ESR Toulouse
ESR INRA - BP52627 - 31326 Castanet Tolosan Cedex GRANCE
[email protected]
05-61-28-50-97
***********************




*
* 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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index