Solved – Zero-inflated negative binomial mixed-effects model in R

count-datamixed modelnegative-binomial-distributionrzero inflation

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!

Best Answer

I think this is the package you need: glmmADMB. I downloaded it here: http://otter-rsch.com/admbre/examples/glmmadmb/glmmADMB.html

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!

Related Solutions

Mixed Model – How to Perform Imputation for a Zero-Inflated Negative Binomial Mixed Effects Model

I suggest that you use predictive mean matching (PMM), the default for continuous data in mice. PMM often works well for semi-continuous data as it makes no distributional assumptions. Check whether the imputed values look plausible (in the sense that they could have been observed had they not been missing). Then, check whether the intra-class correlation of the complete cases is similar to that calculated from the imputed data. If that is the case, I would say that you're done.

BTW: Do the missing data occur only in the outcome? If that's the case, then you might try to stick to the mixed model and not do any imputations at all.

R – Penalized Regression with Zero-Inflated Models

I was looking for a ZIP model as well, and had trouble with mpath. I ended up coding my solutions in Matlab and it seems to work well, but I think it would be very easy in R (provided it is correct/ok). I may have overlooked something so let me know if you see any problems. My data has 80 obs and 550 features.

Recall that ZIP (and ZINB) can be solved via the EM algorithm. I am in the process of implementing this for ZINB, but my ZIP solution is as follows.

My understanding of EM -

Until Convergence: 
   step 1: calc expectations based on parameters
   step 2: maximize with respect to parameters.

To do this in a Lasso framework, set your lambda sequences for Poisson model and the binomial model, call them I and J respectively

for i in I:
    for j in J:
    Until Convergence:
        step 1: calc expectations (estimate latent variable, call it zhat)
        step 2: estimate poisson regression with 1-zhat as weight.  
           estimate logistic regression with zhat as dv.

The issue with step 2 - glmnet wants a factor or two columns to specify the proportion. Maybe just multiply the zhat by 100 and use that appropriately.

I picked i and j via grid search. Additionally, I do not go the full way down the sequence (maybe 30 on both). I also added in a break similar to the GLMNET package to stop before saturation.

Hope this helps.

Best Answer

Related Solutions

Mixed Model – How to Perform Imputation for a Zero-Inflated Negative Binomial Mixed Effects Model

R – Penalized Regression with Zero-Inflated Models

Related Question