poisgof {epiDisplay}

Goodness of fit test for modeling of count data

Description

Poisson and negative binomial regression are used for modeling count data. This command tests the deviance against the degrees of freedom in the model thus determining whether there is overdispersion.

Usage

poisgof(model)

Arguments

model

A Poisson or negative binomial model

Details

To test the significance of overdispersion of the errors of a Poisson or negative binomial model, the deviance is tested against degrees of freedom using chi-squared distribution. A low P value indicates significant overdispersion.

Author(s)

Virasakdi Chongsuvivatwong [e-mail]

See Also

‘glm’

Examples

library(MASS)
quine.pois <- glm(Days ~ Sex/(Age + Eth*Lrn), data = quine, family=poisson)
poisgof(quine.pois)
quine.nb1 <- glm.nb(Days ~ Sex/(Age + Eth*Lrn), data = quine)
poisgof(quine.nb1)