# Zero-inflated negative binomial mixed-effects model in R

Is there such a package that provides for zero-inflated negative binomial mixed-effects model estimation in R?

By that I mean:

• Zero-inflation where you can specify the binomial model for zero inflation, like in function zeroinfl in package pscl:

zeroinfl(y~X|Z, dist = "negbin")
where Z is the formula for the zero inflation model;

• Negative binomial distribution for the count part of the model;

• Random effects specified similar to function lmer of package lme4.

I understand glmmADMB can do all that, except the formula for zero inflation cannot be specified (it is just an intercept, i.e. Z is just 1). But are there any other packages that can do it all?

I will be very thankful for your help!

• I'm looking for the same thing. I think glmmADMB is what I want, but I can't get it running. Dec 7, 2012 at 20:41
• I'm wondering whether or not the ZIM or aod packages can do what you want to do? Jun 26, 2013 at 10:28
• As an update, the glmmTMB package by Ben Bolker supports a zero-inflated generalized linear mixed model (ZIGLMM). Feb 22, 2018 at 21:55

But I still had some problems to get it to run, so I followed the instructions provided in this link and now it works fine http://glmmadmb.r-forge.r-project.org/

Hope this helps!

• Note that this package only allows the fitting of a constant term for the zero-inflation part of the model. Apr 19, 2018 at 16:19

The pscl package provides for a zero inflated Poisson model. I don't think that it can do a negative binomial model, but it might be a place to start. The linked JSS article also discusses related packages, which may lead you to what you're looking for.

• The pscl package does (now) allow negative binomial models by using zeroinfl(..., dist = "negbin", ...) Apr 19, 2018 at 16:20

Depending on what you're trying to do, you might want to look at the aster package. Aster models allow joint analysis of multiple variables that have different probability distributions, and recently have been updated to allow for random effects. They were designed for life history analysis and will work in situations where you can split your response into distinct parts with different distributions, (e.g. survival = Bernoulli, reproduction = Poisson). They can handle "zero-inflation" by modeling the majority of zeroes as bernoulli, and the remainder of the response as negative binomial.

You'll find plenty of documentation here:

http://www.stat.umn.edu/geyer/aster/

There are a lot of packages available. This link might be helpful: https://journal.r-project.org/archive/2017/RJ-2017-066/RJ-2017-066.pdf