EUVIMEDEUVIMED entdecken
Crossref

Changepoint Detection in Multivariate Climate Time Series: Performance Assessment of a Hybrid PELT and Isolation Forest Approach Against Baseline PELT

Adebola Abimbola Ademuwagun, Haruna Umar Yahaya, Samuel Olorunfemi Adams

Faculty of Natural and Applied Sciences Journal of Mathematical and Statistical Computing · 31.03.2026 · 3 (1) · 1-14

DOI: 10.63561/jmsc.v3i1.1207 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Faculty of Natural and Applied Sciences Journal of Mathematical and Statistical Computing, online erschienen am 31.03.2026. Verfasst von Adebola Abimbola Ademuwagun und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The research studied the performance of Hybrid Pruned Exact Linear Time and Isolation Forest (PELT + I Forest) against Baseline PELT in accurately detecting change points in Climate time series data using simulation. The research adopts a Monte Carlo simulation framework to design and evaluate a hybrid change-point detection technique that combines the Pruned Exact Linear Time (PELT) algorithm with machine learning anomaly detection method (I Forest). The hybrid approach (PELT+I Forest) is compared against baseline PELT using simulated multivariate climate datasets. Across small, moderate, and large sample sizes, the same directional patterns persist, temperature and humidity increase while rainfall decreases with more breaks. However, larger samples make the regime shifts more distinct and less noisy. This finding underscores that robust detection methods must perform well not only in large datasets but also in small samples, where noisy signals make breaks harder to capture. Enhanced detection algorithm like PELT+I Forest is therefore vital for early warning in short observational records or regional climate data with limited length. PELT+I Forest delivered consistent performance across moderate and large sample sizes, with marginal but steady improvements in balanced accuracy, making it more robust for longer or multivariate series.This study carried out a Performance Assessment on the developed, implemented and evaluated Hybrid change-point detection framework which integrates the Pruned Exact Linear Time (PELT) algorithm with machine learning anomaly detection method (Isolation Forest) for robust identification of structural breaks in multivariate climate time series against baseline PELT Algorithm. Relative to all other alternatives for Change Point Detection the hybrid PELT+I Forest approach balanced computational efficiency, interpretability, and robustness which retained PELT’s exact segmentation while using ML scorers to capture multivariate, nonlinear anomalies.

Zitieren

APA

Ademuwagun, A. A., Yahaya, H. U., & Adams, S. O. (2026). Changepoint Detection in Multivariate Climate Time Series: Performance Assessment of a Hybrid PELT and Isolation Forest Approach Against Baseline PELT. Faculty of Natural and Applied Sciences Journal of Mathematical and Statistical Computing, 3(1), 1-14. https://doi.org/10.63561/jmsc.v3i1.1207

Vancouver

Ademuwagun AA, Yahaya HU, Adams SO. Changepoint Detection in Multivariate Climate Time Series: Performance Assessment of a Hybrid PELT and Isolation Forest Approach Against Baseline PELT. Faculty of Natural and Applied Sciences Journal of Mathematical and Statistical Computing. 2026;3(1):1-14. doi:10.63561/jmsc.v3i1.1207

BibTeX
@article{Ademuwagun2026,
  author = {Ademuwagun, Adebola Abimbola and Yahaya, Haruna Umar and Adams, Samuel Olorunfemi},
  title = {Changepoint Detection in Multivariate Climate Time Series: Performance Assessment of a Hybrid PELT and Isolation Forest Approach Against Baseline PELT},
  journal = {Faculty of Natural and Applied Sciences Journal of Mathematical and Statistical Computing},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {1--14},
  doi = {10.63561/jmsc.v3i1.1207}
}
Kostenlos starten →