Statalist


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

Re: st: IIA test after svy:mlogit


From   Claudio Cruz Cazares <Claudio.Cruz@uab.cat>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: IIA test after svy:mlogit
Date   Wed, 01 Oct 2008 11:35:14 +0200

Hi, try the  Small-Hsiao Test.

First you have to set a random value:

. set seed 12345

then, 

. mlogtest, smhsiao base

You can try different random values until get the IIA assumption.

Greetings,

Claudio



----- Mensaje original -----
De: "Joao Ricardo F. Lima" <jricardofl@gmail.com>
Fecha: Martes, Septiembre 30, 2008 10:40 pm
Asunto: st: IIA test after svy:mlogit

> Dear Stephen and Statalisters,
> 
> Can I do the Hausman's test for independence of irrelevant
> alternatives after svy:mlogit? My dependent variable has 4
> alternatives.
> Iīm using Stata 10.1.
> 
> I tried to use the -mlogtest-
> 
> STB-58  sg155 . . . . . . . . . . . . .  Tests for the multinomial 
> logit model
>        (help mlogtest, _perhs, _pecats if installed) . J. Freese 
> & J. S. Long
>        11/00   pp.19--25; STB Reprints Vol 10, pp.247--255
>        simplifies the obtaining of several tests associated with
>        the multinomial logit model
> 
> However, you can see below that I canīt use with survey data.
> 
> . mlogtest, hausman base
> 
> Problem determining number of categories.
> 
> **** Hausman tests of IIA assumption
> 
> Ho: Odds(Outcome-J vs Outcome-K) are independent of other 
> alternatives.hausman cannot be used with clustered or p-weighted data
> r(198);
> 
> end of do-file
> 
> r(198);
> 
> Thanks for the help!
> 
> Best Regards,
> 
> Joao Lima
> 
> *
> *   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/
> 


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index