Bookmark and Share

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: checking over dispersion in XTPOISSON


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: checking over dispersion in XTPOISSON
Date   Fri, 6 Jan 2012 10:24:48 +0000

I mentioned -xtnbreg- because no one had mentioned it. I have never
used it. It should be easy enough to check whether a negative binomial
model gives much better fit to the data than a Poisson model. For
example, examine observed and fitted for both models.

I don't know whether that can be encapsulated in a single test or
figure of merit, if that is what you seek.

Nick

On Fri, Jan 6, 2012 at 10:12 AM, Rajaram Subramanian Potty
<rajara999@gmail.com> wrote:
> The out put from "xtnbreg" gives r and s. But how to check whether
> there is overdispersion from r and s. Manual did not explaing how to
> check overdispersion using "xtnbreg". But for "nbreg" the alpha value
> will indicate whether there is overdispersion.
>
> Thanks and regards,
>
> RAJARAM. S
>
> On Fri, Jan 6, 2012 at 2:19 PM, Nick Cox <njcoxstata@gmail.com> wrote:
>> Also, Stata [sic] has -xtnbreg-.
>>
>> Nick
>>
>> On Fri, Jan 6, 2012 at 5:58 AM, Muhammad Anees <anees@aneconomist.com> wrote:
>>
>>> Extending my previous discussion, then
>>>
>>> if it is over or underdispersed. Then apply the cluster option as
>>> shows above. See if the standard errors change much. If change, then
>>> there is not overdispersion
>>
>> On Fri, Jan 6, 2012 at 10:45 AM, Rajaram Subramanian Potty
>>
>>>> Thanks for the suggestion
>>>>
>>>> I understand from the STATA manual that negative binomial regression
>>>> can be used to check overdispersion. I am nost sure whether same can
>>>> be used if one has panel data. Can I use negative binomial regression
>>>> with the cluster option to test the over dispersion in the present
>>>> problem.  In the cluster I will consider the panel identifier.
>>
>> On Thu, Jan 5, 2012 at 1:59 PM, Muhammad Anees <anees@aneconomist.com> wrote:
>>
>>>>> The following quotation from my discussion with Jo Hilbe describes and
>>>>> answers many such problems.
>>>>>
>>>>> glm panelcount x1 x2 x3, fam(poi)
>>>>> glm panelcount x1 x2 x3, fam(poi) cluster(panelvar)
>>>>> glm panelcount x1 x2 x3, fam(nb ml)
>>>>> glm panelcount x1 x2 x3, fam(nb ml) cluster(panelvar)
>>>>>
>>>>> where panelvar is your panel variable, eg id and x1-x3 are predictors.
>>>>>
>>>>> Check the Pearson dispersion statistic for the Poisson model. If it is
>>>>> above 1.0 the model is Poisson overdispersed.
>>
>> Thu, Jan 5, 2012 at 1:21 PM, Rajaram Subramanian Potty
>>
>>>>>> I would like to know how one should check for overdispersion in a
>>>>>> model fitted using "xtpoisson". If one noticed overdispersion how to
>>>>>> correct it or which is the alternative analysis.

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index