Multiple change-point detection for Poisson processes
Abstract
The aim of change-point detection is to identify behavioral shifts within time series data. This article focuses on scenarios where the data are derived from a heterogeneous Poisson process or a marked Poisson process. We present a methodology for detecting multiple offline change-points using a minimum contrast estimator. Specifically, we address how to manage the continuous nature of the process given the available discrete observations. Additionally, we select the appropriate number of changes via a cross-validation procedure, which is particularly effective given the characteristics of the Poisson process. Lastly, we show how to use this methodology for self-exciting processes with changes in the intensity. Through experiments, with both simulated and real datasets, we showcase the advantages of the proposed method, which has been implemented in the R package.
Related
- Laplacian U-Processes for Multiple Change-Point Detection in Dependent Text Networks: An Application to Historical Chinese Articles
- Multiple change-point detection via bottom-up scanning
- Multiple Change Point Detection in the Mean and Covariance of High-Dimensional Matrix Time Series
- Strong Approximations for Multivariate Integrated Empirical Processes: Two-Sample Testing, Change-Point Detection, and Inference with Estimated Parameters
- Exact Multiple Change-Point Detection Via Smallest Valid Partitioning
BibTeX
@article{dionblanc2026multiple,
title = {{Multiple change-point detection for Poisson processes}},
author = {Charlotte Dion-Blanc and Diala Hawat and Émilie Lebarbier and Stephane S. Robin},
journal = {Computational Statistics and Data Analysis},
year = {2026},
month = aug,
doi = {10.1016/j.csda.2026.108464},
eprint = {2302.09103},
archivePrefix = {arXiv},
url = {https://hal.science/hal-05732909v1},
note = {CC BY},
}
RIS
TY - JOUR
TI - Multiple change-point detection for Poisson processes
AU - Charlotte Dion-Blanc
AU - Diala Hawat
AU - Émilie Lebarbier
AU - Stephane S. Robin
PY - 2026
DA - 2026/08/31
JO - Computational Statistics and Data Analysis
DO - 10.1016/j.csda.2026.108464
UR - https://hal.science/hal-05732909v1
AB - The aim of change-point detection is to identify behavioral shifts within time series data. This article focuses on scenarios where the data are derived from a heterogeneous Poisson process or a marked Poisson process. We present a methodology for detecting multiple offline change-points using a minimum contrast estimator. Specifically, we address how to manage the continuous nature of the process given the available discrete observations. Additionally, we select the appropriate number of changes via a cross-validation procedure, which is particularly effective given the characteristics of the Poisson process. Lastly, we show how to use this methodology for self-exciting processes with changes in the intensity. Through experiments, with both simulated and real datasets, we showcase the advantages of the proposed method, which has been implemented in the R package.
SN - 0167-9473
ER -