mRm: An R package for conditional maximum likelihood estimation in mixed Rasch models

Conditional maximum likelihood estimation via the EM algorithm and information-criterion-based model selection in binary mixed Rasch models.

Version: 1.1.5
Published: 2013-12-02
Author: David Preinerstorfer
Maintainer: David Preinerstorfer <david.preinerstorfer at univie.ac.at>
License: GPL-2
NeedsCompilation: yes
In views: Psychometrics
CRAN checks: mRm results

Downloads:

Reference manual: mRm.pdf
Package source: mRm_1.1.5.tar.gz
Windows binaries: r-devel: mRm_1.1.5.zip, r-release: mRm_1.1.5.zip, r-oldrel: mRm_1.1.5.zip
OS X Mavericks binaries: r-release: mRm_1.1.5.tgz, r-oldrel: mRm_1.1.5.tgz
Old sources: mRm archive

Reverse dependencies:

Reverse suggests: psychomix

Linking:

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