Statalist


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

RE: st: A model approach question?


From   jverkuilen <jverkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: A model approach question?
Date   Sun, 29 Mar 2009 13:20:38 -0400

Brad, 

There are a number of models like that, but which one would fit your situation I can't say exactly. Probably they can be estimated using an appropriately constrained -gologit-. 

Alan Agresti has more than one review paper on ordinal models. His webpage should have the exact sites (and reprints?). 

Such models are also discussed in:

L. Fahrmeir, G. Tutz, & W. Hennevogl. 2001. Multivariate Statistical Modelling Based on Generalized Linear Models, 2nd Edition. Springer. 

Gerhard Tutz created several models like that but one of my students has my copy so I can't check.  

-----Original Message-----
From: "Anderson, Bradley" <BJAnderson@butler.org>
To: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Sent: 3/26/2009 1:33 PM
Subject: st: A model approach question?

Apologies if this is naïve but I'm having trouble finding a specific answer.  I'm interesting in modeling a pattern of risk that can be conceptualized as a series of decisions.  It would look something like:

Step1: No-risk behavior versus risk behavior.

        Step2 if risk in Step1:  Risk Type = 1 or 2;

                Step3 if Risk Type = 1:  Low Risk versus High risk.
                Step3 if Risk Type = 2:  Low Risk versus High risk.

So there are basically 5 categorical outcomes across which the probability will sum to 1.0.  But the type of risk taken is conditional on the decision to engage in risky behavior at Step 1.  And for each type of risk there are two levels of risk.  Clearly this can be approached as a multinomial logit model but is there an approach that better captures this type of decision making process?  I think I've seen references to such models but ...  Additionally, I'll need to adjust for within subject clustering.

Brad

This e-mail and any files transmitted with it are confidential and intended solely for the use of the individual
or entity to whom they are addressed.  If you are not the intended recipient, you are hereby notified
that any disclosure, copying, distribution or taking of any action in reliance on the information contained in
this e-mail is prohibited.  If you have received this e-mail in error, please notify sender by reply e-mail and
delete this message and any attachment(s) immediately.  Thank you for your consideration in this matter.

*
*   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/


*
*   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