Re: st: mlogit problems

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

At 09:16 PM 7/29/2006, JOSHUA HAWLEY wrote:
> Hello
>
> I'm running a series of svy mlogit commands on different countries. 
> I'm having some trouble producing the estimates with the expected 
> groupings. In other words as part of the model I have five 
> categories (call them A, B, C, D, E). When I run a tabulation on the 
> dep var I get frequencies as I would expect, 2000 for one, 60 for 
> another, you get the idea, and the important detail is there are 
> valid cases in all five categories. However, when I run the mlogit 
> commands I get a nonsense response back with regression output that 
> seems to show the dep variable as one outcomes. In other words there 
> is just one set of coefficents rather then the four I would expect.
>
> I only get this response on selected datasets, and the same command 
> language works fine on a number of nations.
It probably would help to see some commands and output.  Also, 
remember that, just because a frequency shows cases in every 
category, that doesn't mean you'll have cases in every category when 
you do the mlogit, e.g. you could lose cases because of missing data 
on one or more of your independent variables.

In any event, I am betting it has something to do with your data 
rather than your commands, but without looking at the data, commands 
and output it is hard to say.


-------------------------------------------
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/