Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: Heckman procedure

From   Jan Wynen <>
To   "" <>
Subject   st: Heckman procedure
Date   Thu, 4 Oct 2012 11:01:49 +0000

Dear Statalist'ers,

I am trying to run the following Heckman procedure whereby;

I have following selection equation:

z= x1+x2+x3 +v

whereby v is the error term

and following structural equation:


whereby e is the error term

z1, z2 and z3 are variables that affect only y (structural equation) and not z (selection). They are only observed when z=1.

Is this a good approach? Or will z1, z2 and z3 be endogenous (will they not be independent of the error term of the selection equation)?

Any help would be much appreciated!



*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index