Generalized Linear Models Extended in R

Overview

The R package glmx provides:

  • Estimation of generalized linear models with extra parameters via glmx, e.g., parametric links, or families with additional parameters (such as negative binomial).

  • Estimation of heteroscedastic binary response models via hetglm using maximum likelihood.

  • Density dpregibon, distribution function ppregibon, quantile function qpregibon, and random generation rpregibon for the Pregibon distribution.

  • Various symmetric and asymmetric parametric links for use as link function for binomial generalized linear models, including gosset, pregibon, loglog, nblogit, angular, among others.

Installation

The stable version of glmx is available on CRAN:

install.packages("glmx")

The latest development version can be installed from R-universe:

install.packages("glmx", repos = "https://zeileis.R-universe.dev")

License

The package is available under the General Public License version 3 or version 2