On Wed, 2009-09-16 at 14:55 -0500, Barney Luttbeg wrote: > Check out the pscl package. > > http://cran.r-project.org/web/packages/pscl/index.html > > I haven't used it, but students in my class have.
I don't think the pscl package can handle random effects in its count models (hurdle() and zeroinfl()). The MCMCglmm package for R has a zeroinflated poisson distribution for model fitting, if you don't need the negative binomial. As far as R goes, the real experts to ask are on the R-sig-mixed-models email list. You can sign up here: https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models The guys behind glmmADMB also frequent that list so you might get help with that package as well. HTH G > > Barney Luttbeg > > On Wed, Sep 16, 2009 at 11:55 AM, James Hodson > <[email protected]>wrote: > > > Dear all, > > > > I've been searching for a while now for a package in R that can perform > > mixed-model zero-inflated negative binomial regression. I am looking to > > analyze pellet count data at the individual plot level, while accounting > > for the fact that pellet plots are nested within sites and thus are not > > strictly independent observations. I have a lot of zero counts, which is > > why I'm interested in a zero-inflated model. I have tried the package > > glmmADMB, but I have had no success in getting it to work with my data. > > Any advice would be welcome. > > > > Thanks, > > > > James Hodson > > [email protected] > > -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Dr. Gavin Simpson [t] +44 (0)20 7679 0522 ECRC, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/ UK. WC1E 6BT. [w] http://www.freshwaters.org.uk %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
