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

st: Simultaneous Estimation of Discrete (Polychotomous) Continous Equation System

From   Tarun Kushwaha <>
Subject   st: Simultaneous Estimation of Discrete (Polychotomous) Continous Equation System
Date   Thu, 4 Oct 2007 09:07:06 -0700 (PDT)

Hi All, 

I need help/ suggestions from you on the following. 

I have system of equations which is as follows. 

a = f(b, c, d, p, q, r)
b = f(a, c, d, s, t, u)
c = f(a, b, d, v, w, x)
d = f(b, c, y, z)

a- d are endogenous
p-z are exogenous
a is POLYCHOTOMOUS choice variable 
b-d, and p-z are continuous variables 

How do I solve this system? 

Amemiya (1978) has suggested Generalized Probit and
Heckman (1978) has suggested Heckman's 2SLS to address
such problems. Maddala (1983, pg 242-245) touch upon
some of these classes of model. Though they all deal
with endogeneity between a DICHOTOMOUS choice and a
continuous linear variable. Omar Keshk has a STATA
procedure cdsimeq to address these types of problems. 
Papers in Health Economics (Hamilton 1997),
Agriculture Economics (Zepeda 1994, Hayes et al 1997),
and Accounting (Copley et al 1994, 1995) have used
Generalized Probit to account for endogeneity in
continuous discrete (dichotomous) models. 

However, I have not been able to track either
theoretical or applied paper in Econometrics or
Statistics that solves endogeneity issue between
POLYCHOTOMOUS choice and Linear variables.

Any suggestions, help, starting point are highly

Thanks in advance


Shape Yahoo! in your own image.  Join our Network Research Panel today! 

*   For searches and help try:

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