glmfitmiss: Fitting GLMs with Missing Data in Both Responses and Covariates

Fits generalized linear models (GLMs) when there is missing data in both the response and categorical covariates. The functions implement likelihood-based methods using the Expectation and Maximization (EM) algorithm and optionally apply Firth’s bias correction for improved inference. See Pradhan, Nychka, and Bandyopadhyay (2025) <https:>, Maiti and Pradhan (2009) <doi:10.1111/j.1541-0420.2008.01186.x>, Maity, Pradhan, and Das (2019) <doi:10.1080/00031305.2017.1407359> for further methodological details.

Version:	2.1.0
Depends:	R (≥ 4.0.0)
Imports:	data.table (≥ 1.12.8), dplyr (≥ 1.0.0), abind (≥ 1.4-5), MASS (≥ 7.3-53), brglm2 (≥ 0.7.1)
Suggests:	testthat (≥ 3.0.0)
Published:	2025-04-22
DOI:	10.32614/CRAN.package.glmfitmiss
Author:	Vivek Pradhan [aut, cre], Douglas Nychka [aut], Soutir Bandyopadhyay [aut]
Maintainer:	Vivek Pradhan <vpradhan2009 at gmail.com>
License:	MIT + file LICENSE
NeedsCompilation:	no
CRAN checks:	glmfitmiss results

Documentation:

Reference manual:

glmfitmiss.pdf

Downloads:

Package source:	glmfitmiss_2.1.0.tar.gz
Windows binaries:	r-devel: not available, r-release: glmfitmiss_2.1.0.zip, r-oldrel: glmfitmiss_2.1.0.zip
macOS binaries:	r-release (arm64): glmfitmiss_2.1.0.tgz, r-oldrel (arm64): glmfitmiss_2.1.0.tgz, r-release (x86_64): glmfitmiss_2.1.0.tgz, r-oldrel (x86_64): glmfitmiss_2.1.0.tgz

Linking:

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