Am Donnerstag, den 15.01.2009, 10:26 -0500 schrieb Johanna Avato:
> I am running an ordered logit model with 3 categories. I am wondering
> if it is better to report the proportional odds ratios (or their
> percentage change) or the marginal effects for each outcome. Any
> suggestions? Usually, I like the marginal effects but I find them less
> intuitive when the sign is actually changing across the outcome
> categories.
I would use plots to show the results of order logit models. Take a
look at Long/Freese (2006: 211).
@BOOK{long06,
author = {Long, J. Scott and Freese, Jeremy},
year = {2006},
title = {Regression Models for Categorical Dependent Variables Using
Stata},
address = {College Station},
publisher = {Stata Press},
edition = {2}
}
--
[email protected]
030 25491-361
*
* 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/
