Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Mixlogit Error
From
Arne Risa Hole <[email protected]>
To
[email protected]
Subject
Re: st: Mixlogit Error
Date
Fri, 2 Aug 2013 11:16:50 +0100
Hi Cameron,
What you need to ask yourself is whether your model can be fit with
-asclogit-, as -mixlogit- is just -asclogit- with random coefficients.
I suspect the answer is no, as the asclogit/mixlogit class of models
assume that a decision-maker chooses a single alternative from a set
of available options. It is possible to have repeated choices, but on
each choice occasion only one alternative is chosen.
-clogit- can be used to estimate a range of models which are
equivalent computationally but very different conceptually.
Arne
On 2 August 2013 09:06, Cameron McLoughlin <[email protected]> wrote:
> Dear Statlisters,
>
> I have a question regarding the correct use of the user written
> command mixlogit.
>
> I have a monthly panel of 27 countries over 7 years and am trying to
> estimate a mixed logit model with a crisis dummy (0/1) as the
> dependent variable. My data is stacked in the usual STATA format, for
> example:
>
>
> Id Time Cisis Var1 Var2
>
> 1 1994m1 0
> 1 1994m2 1
> 1 1994m3 0
>
>
> Following previous advice on this forum available in other FAQs
> regarding mixed logit, I first made sure that I can estimate a clogit
> model using this data. I have no problems using for example with the
> command:
>
> clogit crisis var1 var2, group(id) vce(cluster id)
>
> After reading Arne Hole's STATA Journal article on the mixed logit
> command, I thought that the way I should code this command should be:
>
> mixlogit crisis, rand(var1 var2) group(time) id(id)
>
> When I try this piece of code though, I receive the error message 'at
> least one group has more than one chosen alternative'.
>
> Any ideas as to what I am doing wrong? I guess that I am not
> specifying the group() option i.e. the choice sets correctly, but I'm
> still strugglin with how to specify this option.....
>
> Thanks alot for any suggestions you may have!!!
> Cameron McLoughlin
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/