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

Re: st: Sample selection variable for bivariate probit selection

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: Sample selection variable for bivariate probit selection
Date   Fri, 25 Jan 2008 13:37:30 +0000 (GMT)

--- Carlos Gamero <[email protected]> wrote:
> Thank you very much for your answer. The only question is that I dont
> think I have two selection equations. I think I have one (do you work
> or not) but it must be estimated using a bivariate probit model,
> because the "being in education" decission is not exogenous, i.e., if
> you do work, then you probably find it harder to study (and vice 
> versa). That's why I would like to estimate the selection equation as
> part of a bivariate probit model and, then, somehow take account of
> the selection problem later in the satisfaction equation (estimated
>  as another probit model).

Ok, I think I understand your question now. However, what you want to
do sounds like overkill to me: you are tyring to correct the
correction. Moreover, both models rely on pretty strong assumptions,
bunching them together only makes that worse. I would just go for 
-heckprob-, but if you really want to do it, than you should write down
the likelihood and maximize it using -ml-. If you want to go down that
road you will probably want to buy . But I really don't think it
is worth the effort.

Hope this helps,

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

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Sent from Yahoo! Mail - a smarter inbox

*   For searches and help try:

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