MXM: Discovering Multiple, Statistically-Equivalent Signatures

Feature selection methods for identifying minimal, statistically-equivalent and equally-predictive feature subsets. Bayesian network algorithms and related functions are also included. The package name 'MXM' stands for "Mens eX Machina", meaning "Mind from the Machine" in Latin.

Version: 0.9.4
Imports: methods, stats, utils, survival, MASS, TunePareto, Hmisc, graphics, ordinal, nnet, betareg, quantreg, pscl, lme4, foreach, doParallel, parallel, speedglm, e1071, relations, ggm, Rfast
Suggests: gRbase, Biobase, hash, Rgraphviz
Published: 2016-08-05
Author: Ioannis Tsamardinos, Vincenzo Lagani, Giorgos Athineou, Michail Tsagris, Giorgos Borboudakis, Anna Roumpelaki
Maintainer: Michail Tsagris <mtsagris at>
BugReports: NA
License: GPL-2
NeedsCompilation: no
Citation: MXM citation info
CRAN checks: MXM results


Reference manual: MXM.pdf
Package source: MXM_0.9.4.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Mavericks binaries: r-release: MXM_0.9.4.tgz, r-oldrel: MXM_0.9.4.tgz
Old sources: MXM archive