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

Re: st: Calculating Marginal Effects for Heckman Probit Model

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: Calculating Marginal Effects for Heckman Probit Model
Date   Tue, 19 Aug 2008 10:16:15 +0100 (BST)

--- [email protected] wrote:
> I am estimating a Heckman Probit model with Selection and some of my
> variables appear in both the outcome and the selection equation.
> My question is: How can I interpret these variables?
> As far as I understand it the coefficients (and marginal effects
> produced
> by Stata) give no indication about the true marginal effect if the
> variable appears in both equations, but so far I have not found a way
> to get the actual marginal effects.

In selection models there are many types of marginal effects possible.
You can estimate all these using -mfx- by using the -predict()- option,
see -help mfx- and -help heckprob postestimation##predict- . I would
not call any of these "true", they are all meaningful representations
of the results in your model. The trick is to understand what these
different results mean and why they are different. A good introduction
can be found in (Breen 1996)

-- Maarten

Richard Breen (1996) "Regression models: Censored, sample selected, or
truncated data", Quantitative Applications in the Social Sciences, nr
111, Thousand Oacks: Sage.

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

Send instant messages to your online friends 
*   For searches and help try:

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