Monday, March 06, 2006

st: slogit versus mlogit

Hi I wonder if anyone can help me? I am trying to model an ordinal response variable with three levels. I do not have proportional odds and so I have been investigating the stereotype model. I understand that if the stereotype model is of its maximum dimension, in my case 2, then the model is equivalent to a multinomial logistic regression model. Given that I believe that a 2-dimensional model would be an improvement over a 1-dimensional model for my data, is there any benefit in fitting the model using slogit rather than mlogit? Alternatively, does anyone know where I can find detailed information on the interpretation of a 2-dimensional stereotype type model from slogit, especially when my explanatory variables are categorical? Thanks Sara

* * For searches and help try: * * *


Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?