User-defined GLM link function

["GIORGI Roch" <roch.giorgi@mail.ap-hm.fr>]

Dear all,

I have to write my own family (a modification of a poisson family) in R in 
order to perform glm analysis and something is vague for me.

My link function is: 
        linkfun <- function(mu, dce=expected.dc) log(mu - dce)
The inverse of the link fnuction is: 
        linkinv <- function(eta, dce=expected.dc) exp(eta)+dce
The variance is the same as this of a possion family: 
        variance <- function(mu) mu

My question concern the mu.eta parameter. It is the same think that the "deriv" 
paramter for the make.family in SPLUS? I don't understand clearly what it mean 
and what is for this family.

Any help will be so much appreciate,

Thanks

Roch


_____________________________________________________
  Roch GIORGI, MD, PhD
  Service de Santé Publique et d'Information Médicale
  Hôpital de la Timone
  264, rue St Pierre 
  13385 Marseille cedex 5 
  Tel: +33 (0)491 384 949
  Fax: +33 (0)491 385 749 
  email: roch.giorgi@ap-hm.fr 
  Web: http://cybertim.timone.univ-mrs.fr/