boundary: Decision Boundary of a Projective Bayes Classifier

In McMiso: Multicore Multivariable Isotonic Regression

Decision Boundary of a Projective Bayes Classifier

Description

Extracts the decision boundary from a fitted projective Bayes classifier. The decision boundary consists of the minimal set of feature combinations that are classified as positive (1), i.e., combinations classified as 1 for which no combination lower in the partial ordering is also classified as 1. Any feature combination at or above a boundary combination in the partial ordering is guaranteed to be classified as 1, making the boundary the most compact representation of the classification rule.

Usage

boundary(object)

Arguments

object

a fitted object of class "pbc", produced by PBclassifier or mcPBclassifier.

Value

A list of class "boundary" containing the following components:

boundary

a numeric matrix of feature combinations forming the decision boundary, where each row is one minimal positive combination. NULL if no combinations are classified as 1.

yhat

a binary numeric vector giving the classification of each boundary combination (all values will be 1)

pt

a numeric vector of posterior probabilities at each boundary combination

n_boundary

an integer giving the number of boundary combinations

n_positive

an integer giving the total number of combinations classified as 1

n_total

an integer giving the total number of unique feature combinations

References

Cheung YK, Diaz KM. Monotone response surface of multi-factor condition: estimation and Bayes classifiers. J R Stat Soc Series B Stat Methodol. 2023 Apr;85(2):497-522. doi: 10.1093/jrsssb/qkad014. Epub 2023 Mar 22. PMID: 38464683; PMCID: PMC10919322.

Cheung YK, Kuhn L. Evaluating multiplex diagnostic test using partially ordered Bayes classifier. Ann Appl Stat. In press.

Examples

A <- as.matrix(expand.grid(rep(list(0:1), 6)))
set.seed(2025)
X <- A[sample(nrow(A), size=500, replace=TRUE),]
y <- as.numeric(rowSums(X) >= 3)
fit <- PBclassifier(X, y)
db <- boundary(fit)
print(db)

McMiso documentation built on April 4, 2026, 1:07 a.m.