jTree: Junction tree

In gRapHD: Efficient Selection of Undirected Graphical Models for High-Dimensional Datasets

Description

Finds a junction tree.

Usage

jTree(model)

Arguments

model

object of gRapHD class.

Details

Returns one possible junction tree. Note that each edge is associated to one separator in the list, and a separator may be contained in other(s) separator(s). To identify which separator is associated to each edge is enough to check ind<-indSepOrig[which(indSepOrig!=1)]. In this way, the edge juncTree[i,] is associated with separator ind[i].

Value

A list with:

separators	list with unique minimal separators.
juncTree	edges in the tree (each vertex is a clique in the list below).
sepSubSetOfSep	list in which each element gives all the separators which contain this respective separator.
indSepOrig	index of the original separator (in the MCS result) in the list above.
cliques	list with cliques.

Author(s)

Gabriel Coelho Goncalves de Abreu ([e-mail])

Examples

edges <- matrix(c(1,2,2,3,2,4,2,5,2,6,3,4,4,
                  5,5,6,7,8,7,9,8,9,8,10,9,10),ncol=2,byrow=TRUE)
m <- new("gRapHD",edges=edges)
jT <- jTree(m)
str(jT)
# List of 5
#  $ separators    :List of 5
#   ..$ : int(0)
#   ..$ : int 2
#   ..$ : int [1:2] 2 4
#   ..$ : int [1:2] 2 5
#   ..$ : int [1:2] 8 9
#  $ juncTree      : int [1:4, 1:2] 1 2 3 5 2 3 4 6
#  $ sepSubSetOfSep:List of 5
#   ..$ : int [1:4] 2 3 4 5
#   ..$ : int [1:2] 3 4
#   ..$ : int(0)
#   ..$ : int(0)
#   ..$ : int(0)
#  $ indSepOrig    : int [1:6] 1 2 3 4 1 5
#  $ cliques       :List of 6
#   ..$ : int [1:2] 1 2
#   ..$ : int [1:3] 2 3 4
#   ..$ : int [1:3] 2 4 5
#   ..$ : int [1:3] 2 5 6
#   ..$ : int [1:3] 7 8 9
#   ..$ : int [1:3] 8 9 10

gRapHD documentation built on Feb. 9, 2018, 6:05 a.m.