emax.glm: General Tools for Building GLM Expectation-Maximization Models

Implementation of Expectation Maximization (EM) regression of general linear models. The package currently supports Poisson and Logistic regression with variable weights, with underlying theory included in the vignettes. New users are recommended to look at the em.glm() and small.em() functions - the outputs of which are supported by AIC(), BIC(), and logLik() calls. Several plot functions have been included for useful diagnostics and model exploration. Methods are based on the theory of Dempster et al (1977, ISBN:00359246), and follow the methods of Hastie et al. (2009) <doi:10.1007/978-0-387-21606-5_7> and A. Zeileis et al (2017) <doi:10.18637/jss.v027.i08>.

Version: 0.1.2
Depends: R (≥ 2.10)
Imports: MASS, pracma, stats, pander, pROC, AER
Suggests: spelling, knitr, rmarkdown
Published: 2019-07-04
Author: Robert M. Cook ORCID iD [aut, cre]
Maintainer: Robert M. Cook <robert.cook at bcu.ac.uk>
License: GPL-3
NeedsCompilation: no
Language: en-US
Materials: README
CRAN checks: emax.glm results


Reference manual: emax.glm.pdf
Vignettes: em.glm vignette
The EM GLM algorithm - Proof of predicted values
Residual Theory
Warm-up and exposure in the EM algorithm
Package source: emax.glm_0.1.2.tar.gz
Windows binaries: r-devel: emax.glm_0.1.2.zip, r-release: emax.glm_0.1.2.zip, r-oldrel: emax.glm_0.1.2.zip
OS X binaries: r-release: emax.glm_0.1.2.tgz, r-oldrel: emax.glm_0.1.2.tgz


Please use the canonical form https://CRAN.R-project.org/package=emax.glm to link to this page.