[R-sig-ME] need help in glm

[Nicholas Lewin-Koh]

Hi,
You will have to create your own family function to fit 
using glm. look at family.glm in the stats package. You might also
check VGAM and gamlss.mx to see if this or something similar is already
implemented.

Nicholas
essage: 4
> Date: Tue, 29 Sep 2009 08:04:59 -0700 (PDT)
> From: hossain zamani <zamani_hossain at yahoo.com>
> Subject: [R-sig-ME] need help in glm
> To: R-SIG-Mixed-Models at R-project.org
> Message-ID: <456079.40196.qm at web32703.mail.mud.yahoo.com>
> Content-Type: text/plain; charset="iso-8859-1"
> 
> Dear all,
> I am a R using. I have a mixed Poisson regression. my function is:
> 
> ?## P(N=k)=[(a+2)/a
> ](lambda^k)*[((a+2/a+1)+lambda)^(-k-1)-(a+2+lambda)^(-k-1)];
> lambda=a0+a1*x1+a2*x2
> 
> ?and I've attached in a pdf file also.This is a poisson mixed with a
> weighted exponential distribution in the case that
> E(theta)=1(theta=random.effect). in fact this model is?same as a
> Poisson-inverse gaussian or Poisson-lognormal regression.I want to
> estimate the parameters by fitting this model on my data and then compare
> with Poisson regression(glm).Which package in R can be used.
> ?with the best wishes
> ?Zamani.H
> 
>