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]
st: RES: RE: RES: RE: Poisson Regression
From
José Maria Pacheco de Souza <[email protected]>
To
<[email protected]>
Subject
st: RES: RE: RES: RE: Poisson Regression
Date
Mon, 14 Feb 2011 20:42:59 -0200
Paul,
Good point. I will visit again yours references.
As far your observation about using -binreg-, I gave up, because of non
convergence. With -poisson- I didn´t have convergence problems, but sometime
probability estimate greater than 1.
Regards,
José maria
José Maria Pacheco de Souza
Professor Titular, aposentado; Colaborador Sênior
Departamento de Epidemiologia/Faculdade de Saúde Pública/Universidade de São
Paulo
Av. Dr. Arnaldo, 715 - São Paulo, Capital - cep 01246-904
Fones: FSP= (11)3061-7747 Res= (11)3714-2403; (11)3768-8612
www.fsp.usp.br/~jmpsouza
-----Mensagem original-----
De: [email protected]
[mailto:[email protected]] Em nome de Visintainer, Paul
Enviada em: segunda-feira, 14 de fevereiro de 2011 19:02
Para: '[email protected]'
Assunto: st: RE: RES: RE: Poisson Regression
Jose,
The McNutt articles (in my previous reply) suggest that the conversion
method can produce biased point estimates and confidence intervals that are
too narrow, particularly in multivariable settings. Note that in his help
file on -oddsrisk-, Hilbe uses Poisson regression with the robust standard
errors as a comparison for the output of his -oddsrisk- program.
-p
________________________________________________
Paul F. Visintainer, PhD
Baystate Medical Center
Springfield, MA 01199
*
* 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/