Statalist The Stata Listserver

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

Re: st: two step heckman test versus ML

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: two step heckman test versus ML
Date   Sun, 8 Oct 2006 22:31:50 +0100 (BST)


You don't tell us much that can help us help you. For instance, it
would be helpful if you told us the exact commands you typed. 

Sometimes -heckman- models get unstable if you have the same variables
in the selection equation as in wage equation. If that is the case,
than that might show up as different results for ml and twostage. The
solution is to find a variable that affects selection but not the wage
(or whatever dependent variable you happen to be using). This is a
major weak point of -heckman- models, since those variables are
extremely hard to find.


--- lobna bouslimi <[email protected]> wrote:
> The two-step and maximum likelihood estimations gave me two very
> different results. The results of the twostep is more reasonable to
> me, as the coefficients are expected. However, I was told that the ML
> is more efficient. The last iteration of the ML was concave.
> What is the problem? How to make the two estimations consistent and
> what is the difference between both results?

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

All new Yahoo! Mail "The new Interface is stunning in its simplicity and ease of use." - PC Magazine
*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index