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: Heteroscedastic logit model
From
Dieter Somers <[email protected]>
To
[email protected]
Subject
Re: st: Heteroscedastic logit model
Date
Fri, 20 May 2011 11:24:35 +0200
> Okay, thanks for your answers!
>
> Regards,
> Dieter
>
>
> On Fri, May 20, 2011 at 10:59 AM, Nick Cox <[email protected]> wrote:
>>
>> 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 <[email protected]> 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/
>
*
* 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/