hcp_binseg: Binary Segmentation (BinSeg)

In harbinger: A Unified Time Series Event Detection Framework

Binary Segmentation (BinSeg)

Description

Multi-change-point detection via Binary Segmentation on mean/variance using the changepoint package.

Usage

hcp_binseg(Q = 2)

Arguments

Q	Integer. Maximum number of change points to search for.

Details

Binary Segmentation recursively partitions the series around the largest detected change until a maximum number of change points or stopping criterion is met. This is a fast heuristic widely used in practice.

Value

hcp_binseg object.

References

• Vostrikova L (1981). Detecting "disorder" in multidimensional random processes. Soviet Mathematics Doklady, 24, 55–59.

• Killick R, Fearnhead P, Eckley IA (2012). Optimal detection of changepoints with a linear computational cost. JASA, 107(500):1590–1598. dplyr::context

Examples

library(daltoolbox)

# Load change-point example data
data(examples_changepoints)

# Use a simple example
dataset <- examples_changepoints$simple
head(dataset)

# Configure the BinSeg detector
model <- hcp_binseg()

# Fit the detector (no-op for BinSeg)
model <- fit(model, dataset$serie)

# Run detection
detection <- detect(model, dataset$serie)

# Show detected change points
print(detection[(detection$event),])

harbinger documentation built on Feb. 11, 2026, 9:06 a.m.