Learn R Programming
⚠️There's a newer version (0.9.7) of this package.Take me there.

MXM (version 0.9.4)

Discovering Multiple, Statistically-Equivalent Signatures

Description

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.

Copy Link

Version

Install

install.packages('MXM')

Monthly Downloads

1,876

Version

0.9.4

License

GPL-2

Maintainer

Michail Tsagris

Last Published

August 5th, 2016

Functions in MXM (0.9.4)

Transformation of a DAG into an essential graph

Transforms a DAG into an essential graph
Forward selection

Variable selection in regression models with forward selection
ROC and area under the curve

ROC and area under the curve
Forward selection with generalised linear regression models

Variable selection in generalised linear regression models with forward selection
BIC based forward selection with generalised linear models

Variable selection in generalised linear models with forward selection based on BIC
BIC based forward selection

Variable selection in regression models with forward selection using BIC
Conditional independence tests for survival data

Conditional independence test for survival data
Backward selection with generalised linear regression models

Variable selection in generalised linear regression models with backward selection
Cross-Validation for SES and MMPC

Cross-Validation for SES and MMPC
Correlation based tests with and without permutation p-value

Fisher conditional independence test for continuous class variables with and without permutation based p-value
MMPC.temporal.output-class

Class "MMPC.temporal.output"
Neighbours of nodes in an undirected graph

Returns and plots, if asked, the node(s) and their neighbour(s), if there are any.
MMPC solution paths for many combinations of hyper-parameters

MMPC solution paths for many combinations of hyper-parameters
The max-min Markov blanket algorithm

Max-min Markov blanket algorithm
Forward selection with linear regression models

Variable selection in linear regression models with forward selection
MXM-package

This is an R package that currently implements feature selection methods for identifying minimal, statistically-equivalent and equally-predictive feature subsets. In addition, two algorithms for constructing the skeleton of a Bayesian network are included.
Skeleton of the max-min hill-climbing (MMHC) algorithm

The skeleton of a Bayesian network as produced by MMHC
MXM-internal

Internal MXM Functions
MMPCoutput-class

Class "MMPCoutput"
Plot of an (un)directed graph

Plot of an (un)directed graph
CondInditional independence tests

MXM Conditional independence tests
Partial correlation between two variables

Partial correlation
Cross-validation for ridge regression

Cross validation for the ridge regression
Skeleton of the PC algorithm

The skeleton of a Bayesian network produced by the PC algorithm
Ridge regression

Ridge regression
Permutation based p-value for the Pearson correlation coefficient

Permutation based p-value for the Pearson correlation coefficient
Regression models fitting

Regression modelling
Orientation rules for the PC algorithm

The orientations part of the PC algorithm.
Ridge regression coefficients plot

Ridge regression
Constraint based feature selection algorithms

SES: Feature selection algorithm for identifying multiple minimal, statistically-equivalent and equally-predictive feature signatures MMPC: Feature selection algorithm for identifying minimal feature subsets
Plot of longitudinal data

Plot of longitudinal data
Conditional independence tests for sucess rates

Binomial regression conditional independence test for success rates (binomial)
Correlation based conditonal independence tests

Fisher and Spearman conditional independence test for continuous class variables
Constraint based feature selection algorithms for longitudinal and clustered data

SES.temporal: Feature selection algorithm for identifying multiple minimal, statistically-equivalent and equally-predictive feature signatures MMPC.temporal: Feature selection algorithm for identifying minimal feature subsets
Regression models based on SES and MMPC outputs

Regression model(s) obtained from SES or MMPC
SESoutput-class

Class "SESoutput"
SES.temporal.output-class

Class "SES.temporal.output"
Conditional independence test for longitudinal and clustered data

Linear mixed models conditional independence test for longitudinal class variables
Conditional independence test for case control data

Conditional independence test based on conditional logistic regression for case control studies
Conditional independence test for proportions/percentages

Beta regression conditional independence test for proportions/percentage class dependent variables and mixed predictors
Conditional independence test for binary, categorical or ordinal data

Conditional independence test for binary, categorical or ordinal class variables
Conditional independence tests for count data

Regression conditional independence test for discrete (counts) class dependent variables
Conditional independence test for continuous, binary and count data with thousands of samples

Conditional independence test for continuous, binary and discrete (counts) variables with thousands of observations
Conditional independence tests for continous univariate and multivariate data

Linear regression conditional independence test for continous univariate and multivariate response variables
Undirected path(s) between two nodes

Undirected path(s) between two nodes