LOGIT (version 1.3)

toRR: Display count model coefficient table as incidence rate ratios and associated statistics.

Description

Following the glm or glm.nb commands, toRR() displays a table of incidence rate ratios and related statistics including exponentiated model confidence intervals.

Usage

toRR(object)

Arguments

object

name of the fitted glm function model

Value

list

Format

object

The only argument is the name of the fitted glm or glm.nb function model

or

incidence rate ratio of predictor

delta

Model standard error using delta method

zscore

z-statistic

pvalue

probability-value based on normal distribution

exp.loci

Exponentialed lower model confidence interval

exp.upci

Expontiated upper model confidence interval

Details

toRR is a post-estimation function, following the use of glm() with the Poisson or negative.binomial families, and following glm.nb().

References

Hilbe, Joseph M. (2015), Practical Guide to Logistic Regression, Chapman & Hall/CRC.

Hilbe, Joseph M. (2014), Modeling Count Data, Cambridge University Press.

See Also

glm, glm.nb

Examples

library(MASS)
library(LOGIT)
data(medpar)
medpar$los<-as.numeric(medpar$los)
mypoi <- glm(los ~ white + hmo + factor(age80), family=poisson, data=medpar)
summary(mypoi)
toRR(mypoi)