Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Interpreting an endogenous switching regime model


From   Bo MacInnis <[email protected]>
To   [email protected]
Subject   st: Interpreting an endogenous switching regime model
Date   Mon, 18 Jul 2005 14:18:20 -0700

Hi,

I have a really basic question that I am embarrassed to ask. But I am rather embarrassed than wrong.

I am estimating an endogenous switching regime model. Instead of stating my esoteric research subject, I use the effect of union status on wage as an example.

To allow both the intercept term and beta coefficients to differ between the union and non-union, I use Heckman selection correction (along with an instrument for union status) and estimate the outcome equation for the union and non-union separately. The Chow test indicates that the betas do differ significantly between the union and non-union in addition to the intercept term. I then conclude that the endogenous switching regime model is appropriate and that the model using -treatreg- that assumes only the intercept term differs is not appropriate.

Under the endogenous switching regime model, I need to answer two questions:
(1) what is the overall effect of union on wage?
---> Is it the difference between the predicted value of wage (including the intercept term) for the union workers and the predicted value of wage for the non-union workers?

(2) what is the effect of union on "return" to region of residence? region of residence is one of the regressors in the outcome equation assumed to be exogenous (I am sorry if the example sounds too contrived).
--> Is it the difference between the beta coefficient of region of residence for the union workers and the beta coefficient of region of residence for the non-union workers?

Thank you for your help!
Bo MacInnis

*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




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