You can try
Ferrari SLP and Cribari-Neto F -Beta regression for modelling rates and
proportions.
I did not trace where it was published but you can download from
http://www.ime.usp.br/~sferrari/beta.pdf
Basilio de Braganca Pereira
Lars Chi Escreveu:
> Hi All,
>
> My dependent variable is a ratio that takes a value of 0 (zero) for 95% of
> the observations and positive non-integer values for the other 5%. What
> model would be appropriate? I'm thinking of fitting a GLM with a Poisson ~.
> Now, becuase it takes non-integer values most statistical softwares (such as
> R) would issue warning messages.
>
> I'll appreciate any thoughts.
>
> Thanks!
>
> Lars.
Basilio de Bragança Pereira
*Titular Professor of Bioestatistics and of Applied Statistics
*FM-School of Medicine and COPPE-Posgraduate School of Engineering and
HUCFF-University Hospital Clementino Fraga Filho.
*UFRJ-Federal University of Rio de Janeiro
*Tel: (55 21) 2562-2594 or /2558/7045
www.po.ufrj.br/basilio/
*MailAddress:
COPPE/UFRJ
Caixa Postal 68507
CEP 21941-972 Rio de Janeiro,RJ
Brasil
|