mgm: Mixed Graphical Model

In causalMGM: Causal Learning of Mixed Graphical Models

Description

Returns the undirected graph of a mixed data set of continuous and discrete variables. This is an improved version of the Lee & Hastie algorithm (JMLR, 2012). The improvements include the use of three sparsity parameters, depending on the edge type (continuous-continuous, continuous-discrete, discrete-discrete) and a subsampling method to find the optimal sparsities. It also outputs the graph to a .txt file

Usage

mgm(ds)

Arguments

ds	DataSet object returned from loadData()

Value

mgm_graph

Graph object, undirected graph resulting from MGM

Author(s)

AJ Sedgewick, Neha Abraham, Panagiotis Benos

References

AJ Sedgewick, I Shi, RM Donovan, PV Benos, "Learning mixed graphical models with separate sparsity parameters and stability-based model selection", 2016, BM Bioinformatics 17(Suppl 5):S175 DOI: 10.1186/s12859-016-1039-0 https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-016-1039-0

Examples

library(rJava)
mgm_init()
ds <- loadSampleData()
mgm_graph <- mgm(ds)

causalMGM documentation built on May 2, 2019, 5:42 a.m.