Statalist


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

Re: st: Multinominal Logit


From   "Xiaoheng Zhang" <zhangx@tcd.ie>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Multinominal Logit
Date   Wed, 22 Aug 2007 14:03:23 +0100

hi Nick,thanks for this.

On 22/08/07, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> Please note: the term "multinominal" is
> a typo. The standard term is "multinomial".
>
> Nick
> n.j.cox@durham.ac.uk
>
> Xiaoheng Zhang
>
> > Sounds like you are going to do a discrete-choice model.A simple
> > suggestion is that if your explanatory variables contain
> > payment-specific characteristics(I guess,such as costs and service
> > providers), you need to use the Conditional Logit Model; if your
> > explanatoiry variables only contain individuals' characteristics, you
> > can use the Multinominal Logit (or the Conditional Logit,it yields the
> > same outcomes as the Mulitinominal Logit will do). My best advice for
> > you to start is to read State manual re. the Multinomial Logit and the
> > Conditional Logit,it provides very good examples on
> > transportation-mode choice and restaurant choice.
>
> Killian Murphy
>
> > > I have survey data on payment instruments and i am
> > attempting to determine
> > > what are the factors that affect an individuals decision to
> > choose one
> > > payment method over another.
> > >
> > > Am i correct in using a multinominal logit model?
> > > Is the command mlogit correct
> > > is it possible that you could provide me with a do.file as
> > an example
>
> *
> *   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/
>
>


-- 
Greetings,

Xiaoheng Zhang (Kevin)
*
*   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/



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