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

Re: st: switching reg without using switchr

From   "Renzo Comolli" <>
To   <>
Subject   Re: st: switching reg without using switchr
Date   Sun, 30 Nov 2003 14:13:50 -0500

Dear Fukugawa-Nobuya

Could you explain a little bit more what you are looking for?
Could you write the equations?

Are you looking for something like 
Lee, Lung-Fei (1978) "Unionism and Wage Rates: A Simultaneous Equations
Model with Qualitative and Limited Dependent Variables", International
Economic Review, Vol. 19(2), pp. 415-433.
which is an extention of Heckman characterized by the following equations
W1=x1*beta1+e1 if S==1
W2=x2*beta2+e2 if S==0
S=(W1-W2)*delta+ z*gamma+e3 
(I am being a bit sloppy with the notation for the latent variable and with
the assumption on the error)
where the S equation is estimated by probit

If so, unfortunately, there is no pre-made program (to my knowledge) to do
this. It is reasonalby easy to do the estimation of the coefficients on your
own thougth, following the explanation in the article given above using 
-by , estimationcommand- 

The standard errors will be wrong though (too small), you have to adjust


*From   <> 
Subject   st: switching reg without using switchr 
Date   Sun, 30 Nov 2003 21:40:40 +0900 (JST) 

Dear statalisters,
I would like to run a switching regression on stata7.
A switchr program downloadable at ssc does not seem to help.
It conducts ols instead of probit for the regime equation and yeilds the
results with the same number of observations for both of the main
equations(equations for regime1 and regime2).
Does anybody know how to conduct switching regression without using switchr?
Thank you in advance.

*   For searches and help try:

© Copyright 1996–2023 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index