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

st: re: nonlinear constrained estimation and "makecns"

From   Kit Baum <>
Subject   st: re: nonlinear constrained estimation and "makecns"
Date   Mon, 28 Apr 2008 07:36:00 -0400

fm514 said

I'm sorry, probably I just made things a bit uclear in the attempt of
simplifying my question too much. I just wanted to focus the attention on
those two equations. I am not sure I can use SUR since I am instrumenting X3
in the model.

Given the model

> Y1=X1*b1+X2*b2+X3*b3
> Y2=X1*t1+X2*t2+X3*t3

If you are 'instrumenting X3' then it is either (a) Y1 or Y2, or (b) you have not written down the full simultaneous system. For X3 to be endogenous in a simultaneous system, you must have an equation in which it appears on the LHS.

Suggest you write down the full system which you are trying to estimate. It may be that the suggested MLE approach (i.e. FIML) is the only way to address this problem, if these equations are truly simultaneous and subject to nonlinear constraints across equations.

Kit Baum, Boston College Economics and DIW Berlin
An Introduction to Modern Econometrics Using Stata:

* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index