Statalist The Stata Listserver


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

st: simultaneous partial observability biprobits


From   Pablo Mitnik <pmitnik@ssc.wisc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: simultaneous partial observability biprobits
Date   Wed, 04 Oct 2006 11:58:45 -0500

Dear all,

I need to fit a system of partial observability bivariate probits. I have two occupational sectors in the economy. I want to model the probability that the unemployed move from unemployment to sector 1, and the probability that the unemployed move from unemployment to sector 2, in one particular year. Each of these probabilities is assumed to be the result of multiplying two probabilities: the probability a worker has of getting a job offer in the corresponding sector, and the probability that the worker getting the offer accepts it. The offer equations have all the same independent variables, but their coefficients are not restricted to be the same, i.e, they may be (and most likely will be) sector specific. The acceptance equation is assumed to be exactly the same regarding offers from either sector, i.e., it is not sector specific. Some of the independent variables in the acceptance equation are not in the offer equations, so identification is possible. Errors across sectors are likely to be correlated.

The command "biprobit" with the option "partial" only works in a situation in which you have one dependent observed variable that is the multiplication of two unobserved variables; the related unofficial commands I found allow more unobserved variables, but don't deal with simultaneous equations either. Hence, it seems to me that there is no Stata command, official or unofficial, that would allow me to do what I want to do (even assuming that errors are not correlated). I am resorting to your collective wisdom to confirm (or, hopefully, disconfirm) that this is really the case.

Also, may be there is some way of doing what I want to do by using some trick on some existing command (instead of writing a ML evaluator from scratch). Any idea would be greatly appreciated.

All best

Pablo Mitnik

--
Pablo A. Mitnik
University of Wisconsin-Madison (http://www.wisc.edu/ )
Department of Sociology ( http://www.ssc.wisc.edu/soc/)
Center on Wisconsin Strategy (http://www.cows.org/ )
1180 Observatory Drive
Room 7114A
Madison, WI 53706
TEL (608) 2621839
E-mail: pmitnik@ssc.wisc.edu
*
* 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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index