[R-sig-ME] How to get dispersion parameter from a binomial mixed model?

v_coudrain at voila.fr v_coudrain at voila.fr
Sat Dec 15 12:35:18 CET 2012

Thank you very much for these useful comments. I run my models with both penalized quasi likelihood and logistic-normal-binomial distribution and I found some 
differences suggesting that penalized quasi likelihood is more conservative, which is something that I already heard. I noticed that logistic-normal-binomial method is 
gaining importance in the recently published papers, so that it seems to be a good opportunity. Anyway it would be very useful to have good support from an in-
depth work-out of this method as you did for Poisson-overdispersion. I'm quite sure that it should appear in the next time. 
Achetez votre tablette numérique au meilleur prix grâce à Voila http://shopping.voila.fr/vitrine/tablettes-pc

More information about the R-sig-mixed-models mailing list