Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

re: st: 3 simultaneous equations

From   Christopher Baum <>
To   "" <>
Subject   re: st: 3 simultaneous equations
Date   Mon, 12 Sep 2011 15:58:14 -0400

I am trying to estimate two-stage least squares simultaneous equations. I
have three endogenous variables Y1, Y2, and Y3 and one instrument for each

Y1 = a0 + a1*Y2 + a2*Y3+a3*X1 + a4*X4 + a5*X5 
Y2 = b0 + b1*Y1 + b2*Y3 + b3*X2 + b4*X4 + b4*X5 
Y3 = c0 + c1*Y1 + c2*Y2 + c3*X3 + c4*X4 + c4*X5

I tried to use xtivreg2 but I am not sure how it works with three equations.
I would really apprecite any hint.

Mark Schaffer's -xtivreg2- (SSC), like official -xtivreg-, is a single-equation estimator for panel data. That is not what you have here. You should use the -reg3- 
command if you are satisfied with the assumption of i.i.d. errors. As a recent posting indicated, you could also use the -gmm- command to estimate this sort
of system with robust or cluster-robust VCE. The -reg3- command cannot do that.

Each of the above equations is overidentified, so you can (and should) examine the overid test stats. Baum et al.'s -overid- on SSC will do that for -reg3-. 
The -gmm- command will produce a Hansen J statistic, which is a test of the overid restrictions.


Kit Baum   |   Boston College Economics & DIW Berlin   |
                              An Introduction to Stata Programming  |
   An Introduction to Modern Econometrics Using Stata  |

*   For searches and help try:

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