search for: equidispersion

It seems that MASS suggest to judge on the basis of sum(residuals(mode,type="pearson"))/df.residual(mode). My question: Is there any rule of thumb of the cutpoiont value? The paper "On the Use of Corrections for Overdispersion" suggests overdispersion exists if the deviance is at least twice the number of degrees of freedom. Are there any further hints? Thanks. -- Ronggui

I have seen several queries about parameterizing the negative binomial scale parameter. This is called the heterogeneous negative binomial. I have written a function called "nbinomial" which is in the msme package on CRAN. Type ?nbinomial to see the help file. The default model is a negative binomial for which the dispersion parameter is directly related to mu, which is how Stata,

The documentation states : An alternative parametrization (often used in ecology) is by the mean ?mu?, and ?size?, the dispersion parameter. However, this fails : > rnbinom(10, mu = 100, size = 0) [1] NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN Warning message: In rnbinom(10, mu = 100, size = 0) : NAs produced For dispersion set to 0, it should work like drawing from a Poisson distribution.

...rd deviations. Cameron and Trivedi (Regression Analysis of Count Data; 1998, p 62-69), provide a formula for easily calculating negative binomial parameter standard errors from maximum likelihood Hessian standard errors (i.e., the standard errors produced via a Poisson regression, assuming equidispersion), but since these are not automatically produced in the glmm.admb output, this is of little help to me. Any help that could be provided with respect to any of my questions would be greatly appreciated. I apologize in advance for the length of this post, but wanted to include all of my quest...