[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: mlogit problem
William Buchanan <email@example.com>
Re: st: mlogit problem
Sun, 17 Feb 2013 17:47:09 -0800
Especially with so few observations, you should really consider a much more parsimonious model. In terms of variable selection, what has the literature in your area found previously? If you have little, or no, theoretical justification for including the variable in the model how would you judge whether a significant relationship is an artifact in your data rather than a truly significant predictor? You could also consider using data reduction techniques to create composite scores of some of your RHS variables. Your question, and the details that you've provided, are really far too broad for any useful advice beyond going back to the drawing board while keeping the principal of parsimony in mind.
Sent from my iPhone
On Feb 17, 2013, at 16:31, saqlain raza <firstname.lastname@example.org> wrote:
> Thanks JVerkuilen for your response. N=360. Yes I want to do variables selection for my study. If this is not a good idea, what should I do?
> Thanks again for your cooperation
> Saqlain RAZA
> PhD Researcher
> ----- Original Message -----
>> From: JVerkuilen (Gmail) <email@example.com>
>> To: firstname.lastname@example.org
>> Sent: Sunday, February 17, 2013 4:07 PM
>> Subject: Re: st: mlogit problem
>> On Sun, Feb 17, 2013 at 6:46 AM, saqlain raza <email@example.com> wrote:
>>> I am trying to fit -mlogit- with aroung 60 covariates (discrete and
>> continous) initially to see the significnat variables. My dependent variable has
>> four alternate choices. Upto 44 covariates it converges. But, if I add one more
>> variable, iteration process starts and at the end, the result is not converged.
>> Any help will be highly acknowleged.
>>> PhD Researcher
>> You need to indicate your N but you are running a probably unfittable
>> model. With 60 covariates in an mlogit model you have 180 parameters
>> and the chances of a non-concave likelihood, collinearity problems or
>> perfect prediction become higher and higher. This is simply going to
>> fail. I'd really go back to rethink your problem as looking at
>> statistical significance is usually not a very good way to do variable
>> selection anyway.
>> * 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/
* For searches and help try: