Re: st: Mlogit with constraints

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 01:34 PM 12/11/2005, narazani wrote:
> Thank you Richard. Here you have all constraints. There are 5 
> alternatives and 2
> independent variables.
>
> Looking forward to hearing to you
> Best, edlira
>
> . cons 1 [2=3]: time cost1  _cons
>
> . cons 2 [3=4]: time cost1   _cons
>
> . cons 3 [4=5]: time cost1   _cons
>
> . mlogit av20 time cost1  if (location>1|location<6|location>6)&purpose==1,
> cons(1-3) basecat(1)
What exactly are you trying to accomplish?  Are you trying to see if 
categories 2, 3, 4, & 5 can all be combined or something?  I am not 
quite sure why the constants come out the way they do, but other than 
that the results seem consistent with your constraints.  You are 
constraining the coefficients to be the same across equations, and 
indeed they are. If that isn't what you want, you need to re-think 
the constraints.

If you are trying to test whether categories can be combined, Long & 
Freese discuss this starting on p.203 of "Regression Models for 
categorical dependent variables using Stata, Revised edition."  Their 
tests do not involve constraints on the intercepts.  Do -findit 
spostado-, install the version that is appropriate for your version 
of Stata, and use the -mlogtest, combine- command.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc 
*
*   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/