mixpoissonreg: Mixed Poisson Regression for Overdispersed Count Data

Fits mixed Poisson regression models (Poisson-Inverse Gaussian or Negative-Binomial) on data sets with response variables being count data. The models can have varying precision parameter, where a linear regression structure (through a link function) is assumed to hold on the precision parameter. The Expectation-Maximization algorithm for both these models (Poisson Inverse Gaussian and Negative Binomial) is an important contribution of this package. Another important feature of this package is the set of functions to perform global and local influence analysis. See Barreto-Souza and Simas (2016) <doi:10.1007/s11222-015-9601-6> for further details.

README.md Analyzing overdispersed count data with the mixpoissonreg package Confidence and prediction intervals with the mixpoissonreg package Global and local influence analysis with the *mixpoissonreg* package Maximum-likelihood estimation with the mixpoissonreg package mixpoissonreg in the tidyverse

Vignettes Man pages API and functions Files

Package details
Author	Alexandre B. Simas [aut, cre] (<https://orcid.org/0000-0003-2562-2829>), Wagner Barreto-Souza [aut] (<https://orcid.org/0000-0003-0831-7881>)
Maintainer	Alexandre B. Simas <alexandre.impa@gmail.com>
License	GPL-2
Version	1.0.0
URL	https://github.com/vpnsctl/mixpoissonreg/ https://vpnsctl.github.io/mixpoissonreg/
Package repository	View on CRAN
Installation	Install the latest version of this package by entering the following in R: `install.packages("mixpoissonreg")`