### Friday, March 03, 2006

## RE: st: Multinomial logit in stata

Albert -

Unless you had repeated observations for the same covariate pattern, I don't see how proportions alone would be enough information to do proper inference. Those same proportions could have arisen from 100, 1000 or 100000 observations. Clearly the inference would be different for those situations. If you did have reps, then trying a beta or dirchlet model fit might work.

Al Feiveson

-----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten buis Sent: Friday, March 03, 2006 2:41 PM To: statalist@hsphsun2.harvard.edu Subject: Re: st: Multinomial logit in stata

Hi Albert: This question does occationally come up in the binomial form (choice between two options), and the answer than is either assuming that the proportion is beta distributed and use -betafit- with the -alternitive- option, or use a GLM approach recommended by Papke and Wooldridge(1996). Now the multivariate equivalent of -betafit- would be -dirichletfit-, though that program has not been written yet. You might have more luck trying to extent the approach of Papke and Wooldridge. HTH, Maarten

Leslie E. Papke and Jeffrey M. Wooldridge, "Econometric Methods for Fractional Response Variables with an Application to 401(k) Plan Participation Rates", Journal of Applied Econometrics, Vol. 11, No. 6, 1996, pp. 619-632.

--- Albert Newton <lacneo@gmail.com> wrote: > Does anyone know if stata could do multinomial logit model with > proportion as dependent variable. > > e.g. > > choice formprop city Lmkrev Lmkstat Lgrowth Lcable > 1 0.22 1.00 5.47 36.00 0.03 0.59 > 2 0.11 1.00 5.47 36.00 0.03 0.54 > 3 0.06 1.00 5.47 36.00 0.03 0.59 > 4 0.06 1.00 5.47 36.00 0.03 0.59 > 5 0.56 1.00 5.47 36.00 0.03 0.56 > 1 0.38 2.00 5.48 39.00 0.04 0.59 > 2 0.22 2.00 5.48 39.00 0.04 0.59 > 3 0.11 2.00 5.48 39.00 0.04 0.59 > 4 0.06 2.00 5.48 39.00 0.04 0.59 > 5 0.23 2.00 5.48 39.00 0.04 0.59

----------------------------------------- between 1/2/2006 and 31/3/2006 I will be visiting the UCLA, during this time the best way to reach me is by email

Maarten L. Buis Department of Social Research Methodology Vrije Universiteit Amsterdam Boelelaan 1081 1081 HV Amsterdam The Netherlands

visiting adress: Buitenveldertselaan 3 (Metropolitan), room Z214

+31 20 5986715

http://home.fsw.vu.nl/m.buis/ -----------------------------------------

___________________________________________________________ Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com * * 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/

* * 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/

Tag: statalist