[ABE-L] Nossos estatísticos

[poliveir em ime.usp.br]

Prezados coelgas

Isso faz lembrar Bertold Brecht em Perguntas de um operário que ler.  
conforme poema logo abaixo, leiam que entenderão por que.

Saudações,

               Paulo Tadeu Oliveira


Citando Basilio De Braganca Pereira <basiliopereira em gmail.com>:

> Nossos estatísticos está fazendo coisas importantes!!!!
> Basílio
>
> Enviado do meu iPhone
>
> Início da mensagem encaminhada
>
>> De: Giulio Flore <giulio.flore em googlemail.com>
>> Data: 2 de junho de 2015 12:09:44 GMT+2
>> Para: medstats em googlegroups.com
>> Assunto: Re: {MEDSTATS} Zero Inflated Cox Regression?
>> Responder A: medstats em googlegroups.com
>>
>> No worries,
>>
>> Good luck with the task
>>
>>> On 2 June 2015 at 12:09, Tzippy Shochat <tz.shochat em gmail.com> wrote:
>>> Thanks Giulio.
>>>
>>>
>>> 2015-06-02 11:33 GMT+03:00 Giulio Flore <giulio.flore em googlemail.com>:
>>>> For what is worth, you might want to have a look at
>>>>
>>>> Rodrigues, J., de Castro, M., Cancho, V. G., & Balakrishnan, N.  
>>>> (2009). COM–Poisson cure rate survival models and an application  
>>>> to a cutaneous melanoma data. Journal of Statistical Planning and  
>>>> Inference, 139(10), 3605-3611. Free download is available on  
>>>> Scholar
>>>>
>>>> The best bit about CMP is that it deals with zero inflated  
>>>> distribution as well as not over-dispered distributions like  
>>>> Poisson.  A seminal article on the subject is
>>>>
>>>> Cordeiro, G. M., Rodrigues, J., & de Castro, M. (2012). The  
>>>> exponential COM-Poisson distribution. Statistical papers, 53(3),  
>>>> 653-664.
>>>>
>>>> It is available from Springer. There was a paper on CMP also on  
>>>> one of the RSS journals.
>>>>
>>>> The added bonus is that CMP regression is available in STATA and in R.
>>>>
>>>> Best of luck,
>>>>
>>>> Giulio
>>>
>>>
>>>> On 2 June 2015 at 09:55, Tzippy Shochat <tz.shochat em gmail.com> wrote:
>>>> Dear All.
>>>>
>>>> A common situation in Cancer research is measuring the Duration  
>>>> of Response, which is measured as the time from clinical response  
>>>> to treatment till the time of disease progression or death.
>>>>
>>>> The problem is that not all patients have a clinical response and  
>>>> there is no clear consensus how to treat these subjects.
>>>>
>>>> One option I've seen used is to do a two part analysis: Logistic  
>>>> Regression for probability of reponse, followed by Cox  
>>>> regression, where non-responders are deleted from the model.
>>>>
>>>> Another option is to assing  a zero duration of  response time to  
>>>> non-responders and treat these patients as censored at time "0".
>>>>
>>>> I found one refference to a Zero Inflated Cox Regression (see  
>>>> below) but I'm not sure if this approach has gained acceptence in  
>>>> medical research.
>>>>
>>>> http://sites.uclouvain.be/IAP-Stat-Phase-V-VI/PhaseVI/publications_2007/TR/TR07034.pdf
>>>>
>>>>
>>>> Any thoughts, examples would be greatly appreciated.
>>>>
>>>> Tzippy Shochat
>>>> --
>>>> --
>>>> To post a new thread to MedStats, send email to  
>>>> MedStats em googlegroups.com .
>>>> MedStats' home page is http://groups.google.com/group/MedStats .
>>>> Rules: http://groups.google.com/group/MedStats/web/medstats-rules
>>>>
>>>> ---
>>>> You received this message because you are subscribed to the  
>>>> Google Groups "MedStats" group.
>>>> To unsubscribe from this group and stop receiving emails from it,  
>>>> send an email to medstats+unsubscribe em googlegroups.com.
>>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>> --
>>> --
>>> To post a new thread to MedStats, send email to MedStats em googlegroups.com .
>>> MedStats' home page is http://groups.google.com/group/MedStats .
>>> Rules: http://groups.google.com/group/MedStats/web/medstats-rules
>>>
>>> ---
>>> You received this message because you are subscribed to the Google  
>>> Groups "MedStats" group.
>>> To unsubscribe from this group and stop receiving emails from it,  
>>> send an email to medstats+unsubscribe em googlegroups.com.
>>> For more options, visit https://groups.google.com/d/optout.
>>> --
>>> --
>>> To post a new thread to MedStats, send email to MedStats em googlegroups.com .
>>> MedStats' home page is http://groups.google.com/group/MedStats .
>>> Rules: http://groups.google.com/group/MedStats/web/medstats-rules
>>>
>>> ---
>>> You received this message because you are subscribed to the Google  
>>> Groups "MedStats" group.
>>> To unsubscribe from this group and stop receiving emails from it,  
>>> send an email to medstats+unsubscribe em googlegroups.com.
>>> For more options, visit https://groups.google.com/d/optout.
>>
>> --
>> --
>> To post a new thread to MedStats, send email to MedStats em googlegroups.com .
>> MedStats' home page is http://groups.google.com/group/MedStats .
>> Rules: http://groups.google.com/group/MedStats/web/medstats-rules
>>
>> ---
>> You received this message because you are subscribed to the Google  
>> Groups "MedStats" group.
>> To unsubscribe from this group and stop receiving emails from it,  
>> send an email to medstats+unsubscribe em googlegroups.com.
>> For more options, visit https://groups.google.com/d/optout.