Change a Negative Binomial fit to a GLM fit

This function modifies an output object from glm.nb() to one that looks like the output from glm() with a negative binomial family. This allows it to be updated keeping the theta parameter fixed.

glm.convert(object)

Arguments

object	An object of class "negbin", typically the output from glm.nb().

Value

An object of class "glm" with negative binomial family. The theta parameter is then fixed at its present estimate.

Details

Convenience function needed to effect some low level changes to the structure of the fitted model object.

See also

glm.nb, negative.binomial, glm

Examples

quine.nb1 <- glm.nb(Days ~ Sex/(Age + Eth*Lrn), data = quine)
quine.nbA <- glm.convert(quine.nb1)
quine.nbB <- update(quine.nb1, . ~ . + Sex:Age:Lrn)
anova(quine.nbA, quine.nbB)
#> Analysis of Deviance Table
#> 
#> Model 1: Days ~ Sex/(Age + Eth * Lrn)
#> Model 2: Days ~ Sex + Sex:Age + Sex:Eth + Sex:Lrn + Sex:Eth:Lrn + Sex:Age:Lrn
#>   Resid. Df Resid. Dev Df Deviance
#> 1       132     167.56            
#> 2       128     166.83  4    0.723