com.fit: Computes COM-Poisson Regression

In compoisson: Conway-Maxwell-Poisson Distribution

Description

Computes the maximum likelihood estimates of the COM-Poisson model for given count data.

Usage

com.fit(x)

Arguments

x	matrix of count data

Details

The argument x should consist of a matrix where the first column is the level and the second column is the count for the corresponding level.

Value

Returns an object containing four fields:

lambda	Estimate of the lambda parameter
nu	Estimate of the nu parameter
z	Normalizing constant
fitted.values	Estimated counts at given levels

Author(s)

Jeffrey Dunn

References

Shmueli, G., Minka, T. P., Kadane, J. B., Borle, S. and Boatwright, P., “A useful distribution for fitting discrete data: Revival of the Conway-Maxwell-Poisson distribution,” J. Royal Statist. Soc., v54, pp. 127-142, 2005.

See Also

com.compute.z, com.loglikelihood

Examples

	data(insurance)
	com.fit(Lemaire);

Example output

Loading required package: MASS
$lambda
[1] 0.09497434

$nu
[1] 0.2720292

$z
[1] 1.102971

$fitted.values
[1] 96987.174132  9211.293330   724.499276    51.033262     3.324158

$log.likelihood
         [,1]
[1,] -36101.1

compoisson documentation built on May 1, 2019, 11:17 p.m.