Thanks Marteen, but that would apply if we were using dummy variables.
When I use effect coding I am expressing the change in the beta
compared to a grand mean of the betas in the population (all
categories combined). This is what is making me think that it is very
unlikely to be an odds, relative odds or odds ratio since the
information from the numerator is used to calculate the reference
value in the denominator.
I think "prevalence rate difference" is what the exp of that beta
represents. How does it sound?
Thanks
Diego
On 2/1/07, Maarten Buis <[email protected]> wrote:
--- Diego Bassani wrote:
> When I estimate two way interactions using xi3 and a mlogit model,
> the rrr option is:
>
> a) an Odds Ratio
> b) a Prevalence rate Difference
> c) a Risk-Density
>
> Since the betas are the deviance from the 'grand mean' of betas I am
> inclined to interpret it as a prevalence rate difference. I am
> seriously doubting myself on this one... any help?
A while ago Bobby Gutierrez wrote an extensive post on Statalist
on the issue of odds ratios or relative risk ratios in
multinomial logistic regression:
http://www.stata.com/statalist/archive/2005-04/msg00678.html
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
--
Diego Bassani
*
* 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/
