Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: simultaneous regression for proportions


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: simultaneous regression for proportions
Date   Tue, 7 Apr 2009 14:26:13 +0000 (GMT)

--- carlo wrote:
> > A quick question: Is there any relationship between
> > the variables? For example, do the proportions add 
> > to for each cross-section unit? 

--- On Tue, 7/4/09, Mike Wazowski wrote:
> they don't 

Then my programs don't apply. These are meant for the 
situation where multiple proportions add up to one.
-fmlogit- exactly implements what Carlo sugested,
and -dirifit- is an alternative approach for the same
problem.

> one variable is an "index of participation" that runs
> from 0 to 1, the other is the "intensity of trust" that
> also runs from 0 to 1 so i have only one row of data
> per subject.  i want to run 
> participation <- trust ....
> 
> but at the same time there is obviously a question of 
> 
> trust <- participation ....

The bigger problem here seems to me identification. 
If I remember correctly you would need an explanatory 
variable that influences participation but not trust
_and_ an explanatory variable that influences trust
but not participation. I find it hard to imagine 
what these variables might be. 

-- Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://home.fsw.vu.nl/m.buis/
-----------------------------------------




      

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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