Why is it then that in binomial and poisson models in R/S+ it is always assumed [tex]\phi = 1[/tex]? If the assumption is not true then the analysis in meaningless, no?

- Thread starter Prometheus
- Start date
- Tags dispersion parameter glm

Why is it then that in binomial and poisson models in R/S+ it is always assumed [tex]\phi = 1[/tex]? If the assumption is not true then the analysis in meaningless, no?