Skip to content
Prev 3908 / 20628 Next

Zero-inflated negative binomial mixed model?

Roberto--

Yau et al. wrote some functions in Splus to fit their models.  At one 
point, I had gotten them (I think from Andy Lee, if memory serves), 
though they are not terribly user-friendly, and I think the last time I 
tried to use them in R (a couple years back), I was unable to do so.

However, I would definitely recommend the MCMCglmm package, which can 
fit an over-dispersed Poisson mixed model.  It includes a 
per-observation random-effect to handle the over-dispersion.

Hope that helps.

cheers, Dave
Roberto wrote:
Dear Listers,

I was wondering if there is any implementation available in R of a GLMM
based on a zero-inflated negative binomial (basically a zero-inflated
negative binomial mixed effects model).
I see at least one paper online (KKW Yau, K Wang, AH Lee - Biometrical
Journal, 2003) where something like this has been developed (but right 
now I
can't read the paper because Wiley Interscience is down for maintenance).

Thanks and best regards,
Roberto Patuelli

********************
Roberto Patuelli, Ph.D.
Istituto Ricerche Economiche (IRE) (Institute for Economic Research)
Universit? della Svizzera Italiana (University of Lugano)
via Maderno 24, CP 4361
CH-6904 Lugano
Switzerland
Phone: +41-(0)58-666-4166
Fax: +39-02-700419665
Email: roberto.patuelli at usi.ch
Homepage: http://www.people.lu.unisi.ch/patuellr