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

st: re: 2sls in case the endogenous is nonlinear

From   Kit Baum <>
Subject   st: re: 2sls in case the endogenous is nonlinear
Date   Sun, 14 Oct 2007 15:52:22 -0400

Carmine said

I would like to estimate a following equation:

y1 = x1 + x1*c + x2 + e1
where: "c": 0,1; and "x1" is explained by x3, x4, x5....

where x1 is an endogenous variable. "ivreg" separates "[varlist1]" of
exogenous variables from "(varlist2 = varlist_iv)" of endogenous and
instruments. However, it is unknown to me the way (the commands) to
implement the dummy variable "c" in the equation. I would find the same
obstacle with a system of equation, in case "x1" would be the other
dependent variable. Could somebody help me with a concrete example? Thanks
very much in advance.

As indicated in many Statalist postings by Mark Schaffer and myself, - ivreg2-, -ivreg- or -ivregrress- is perfectly capable of generating consistent estimates of this equation's parameters. You could also use something like -cdsimeq- (findit cdsimeq-) but that would place a specific form on the auxiliary equation.

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