Estimation and inference from generalized linear models based on various methods for bias reduction. The 'brglmFit' fitting method can achieve reduction of estimation bias by solving either the mean biasreducing adjusted score equations in Firth (1993) and Kosmidis and Firth (2009) , or the median biasreduction adjusted score equations in Kenne et al. (2016) , or through the direct subtraction of an estimate of the bias of the maximum likelihood estimator from the maximum likelihood estimates as in Cordeiro and McCullagh (1991) . Estimation in all cases takes place via a quasi Fisher scoring algorithm, and S3 methods for the construction of of confidence intervals for the reducedbias estimates are provided. In the special case of generalized linear models for binomial and multinomial responses, the adjusted score approaches return estimates with improved frequentist properties, that are also always finite, even in cases where the maximum likelihood estimates are infinite (e.g. complete and quasicomplete separation). 'brglm2' also provides prefit and postfit methods for detecting separation and infinite maximum likelihood estimates in binomial response generalized linear models.

Maintainer  
License  GPL3 
Version  0.1.8.900 
URL 
https://github.com/ikosmidis/brglm2

Package repository  View on GitHub 
Installation 
Install the latest version of this package by entering the following in R:
install.packages("devtools")
library(devtools)
install_github("ikosmidis/brglm2")

ikosmidis/brglm2 documentation built on July 5, 2018, 4:22 a.m.