knitr::opts_chunk$set(echo = TRUE)

multicmp

multicmp is a toolkit containing statistical analysis models motivated by multivariate forms of the Conway-Maxwell-Poisson (COM-Poisson) distribution for flexible modeling of multivariate count data, especially in the presence of data dispersion, via the bivariate COM-Poisson distribution described in Sellers et al. (2016). Currently the package only supports bivariate data. Future development will extend the package to higher-dimensional data.

Note that multicmp references bivpois package (Karlis and Ntzoufras).

To use multicmp, one will first need to install the following two packages:

```R install.packages("numDeriv") install.packages("stats")

One can install the latest released version of multicmp from CRAN with:

```R
  install.packages("multicmp")

Using multicmp

To get started with multicmp right away, see the parameter estimation below. For a more detailed and technical description of the bivariate COM-Poisson distribution, see Sellers et al. (2016).

The multicmp package houses the accidents data set (Arbous and Kerrich, 1951)

```r

helper loading to be hidden

library(numDeriv) load("data/accidents.rdata") source('R/multicmpests.R')

```R
  data(accidents)

```r

multicmpests(accidents)

`````



diagdavenport/multicmp documentation built on May 15, 2019, 8:22 a.m.