Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: xi3: mlogit effect coding

From   "Diego Bassani" <>
Subject   Re: st: RE: xi3: mlogit effect coding
Date   Thu, 1 Feb 2007 12:23:15 -0500

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?



On 2/1/07, Maarten Buis <> 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:

Hope this helps,

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

*   For searches and help try:

Diego Bassani
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index