Statalist The Stata Listserver


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

st: Ordinal logit with sample selection


From   Jenny Säve-Söderbergh <Jenny.Save-Soderbergh@sofi.su.se>
To   statalist@hsphsun2.harvard.edu
Subject   st: Ordinal logit with sample selection
Date   Thu, 19 Oct 2006 15:31:53 +0200 (CEST)

Dear All, 

I have a problem which I really could use some help on! 

I have data on individuals either selecting funds for their portfolio or not 
selecting funds(1 /0 variable). For those who did select funds, I have an 
ordinal variable representing the choice of a high, medium or low risk fund 
(risk=1,2,3). I am interested only in the choice of portfolio risk, but I want 
to control for the possible selection arising from not everyone selecting 
funds. Therefore I want to simultaneously estimate an ordinal logit model and a 
Heckman selection model.  

My data looks like 

Id		First choice			Second choice

1			0				.
2			1				LOW=1
3			1				HIGH=3
4			0				.
5			1				MEDIUM=2		

Would anyone know how I can do this? Would it be possible just to do a probit 
equation on the first choice, calculate the Mills ratio, and then include these 
in the ordinal probit? 

I would greatly appreciate any suggestions. 

Jenny Säve-Söderbergh




*
*   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