[ABE-L] Nossos estatísticos

Ter Jun 2 19:34:28 -03 2015

Citando 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
>
Perguntas de um operário que lê

Bertold Brecht

Quem construiu Tebas, a das sete portas?
Nos livros vem o nome dos reis,
Mas foram os reis que transportaram as pedras?
Babilònia, tantas vezes destruida,
Quem outras tantas a reconstruiu? Em que casas
Da Lima Dourada moravam seus obreiros?
No dia em que ficou pronta a Muralha da China para onde
Foram os seus pedreiros? A grande Roma
Está cheia de arcos de triunfo. Quem os ergueu? Sobre quem
Triunfaram os Césares? A tão cantada Bizâncio
Sò tinha palácios
Para os seus habitantes? Até a legendária Atlântida
Na noite em que o mar a engoliu
Viu afogados gritar por seus escravos.

O jovem Alexandre conquistou as Indias
Sòzinho?
César venceu os gauleses.
Nem sequer tinha um cozinheiro ao seu serviço?
Quando a sua armada se afundou Filipe de Espanha
Chorou. E ninguém mais?
Frederico II ganhou a guerra dos sete anos
Quem mais a ganhou?

Em cada página uma vitòria.
Quem cozinhava os festins?
Em cada década um grande homem.
Quem pagava as despesas?

Tantas histórias
Quantas perguntas

> 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.