At 09:49 AM 3/23/2005 -0500, Zhu, Carolyn wrote:
Hi,
Would any one tell me after estimating ordered logit models which the
proportional odds assumptions are pretty much satisfied, should I still
compute the marginal effects? Also, what is the general rule on the number
of categories the dependent variables can take on?
Thanks,
First off, I would recommend installing Long and Freese's -spostado-
routines, as they include some nice commands for estimating marginal and
discrete effects.
I'm not sure why you wouldn't estimate marginal effects. If you find the
information useful, go ahead.
I don't know that there is any specific rule on number of categories for
the DV. At some point, you might wonder if you have enough categories to
warrant treating the variable as continuous rather than ordinal. I suppose
you might run into potential problems if you have so many categories that
the Ns within a category are very small.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: [email protected]
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
*
* 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/
