Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Heteroscedastic logit model


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Heteroscedastic logit model
Date   Fri, 20 May 2011 09:59:05 +0100

Yes, but that doesn't mean that nothing can be done by way of thinking
about this.

Also, why this cultural assumption that every question of model
assessment and scrutiny is a testing question?

Why not calculate your residuals and then inspect for
heteroscedasticity and think about what it means?

Nick

On Fri, May 20, 2011 at 9:52 AM, Maarten Buis <maartenlbuis@gmail.com> wrote:
> On Fri, May 20, 2011 at 10:08 AM, Dieter Somers wrote:
>> I was also wondering whether there exists a general test that checks
>> for heteroscedasticity in a conditional logit model (for instance a
>> command similar to the 'hettest') ?
>
> Heteroskedasticity is a much biger problem and much harder to check in
> non-linear models than in linear regression, so I think that no such
> test exists or even can exist. At some point you have to live with the
> fact that your model contains assumptions that cannot be tested. For
> more on this, see:
> http://www.stata.com/statalist/archive/2010-11/msg00996.html
>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index