EquivalenceClass: Simplifies H-representation by exploiting symmetry

In MCARtest: Optimal Nonparametric Testing of Missing Completely at Random

Simplifies H-representation by exploiting symmetry

Description

The marginal polytope and related objects have many symmetries. By relabelling the levels of discrete variables we transform facets into other facets. This function reduces a list of halfspace normals to its equivalence classes.

Usage

EquivalenceClass(bS, M, Hrep)

Arguments

bS	A binary matrix specifying the set of observation patterns. Each row encodes a single pattern.
M	A vector of positive integers giving the alphabet sizes of the discrete variables.
Hrep	An H-representation generated by MargPolyHrep, ConsMinkSumHrep or InconsMinkSumHrep.

Value

A list of representative halfspace normals.

Examples

bS=matrix(c(1,1,0, 1,0,1, 0,1,1),byrow=TRUE,ncol=3) # Our canonical 3d example
Hrep=MargPolyHrep(bS,c(2,2,2))
EquivalenceClass(bS,c(2,2,2),Hrep)

MCARtest documentation built on Oct. 29, 2024, 5:08 p.m.