Information technologies and computer sciences

Method of Structural Analysis of Heteroscedastic Time Series Using the Boosting EGARCH Model

Authors

V. Ye. Kopniak
Vinnytsia National Technical University ROR
V. B. Mokin
Vinnytsia National Technical University ROR
DOI https://doi.org/10.31649/1997-9266-2025-183-6-139-148

Keywords

time series data analysis system analysis heteroskedastic model intelligent model generative artificial intelligence air quality

Abstract

A comprehensive method for the structural analysis of a heteroscedastic time series is developed in this study. The approach combines an original boosting-based model of the conditional mean with an EGARCH specification of the conditional variance. The method is designed to quantitatively localize and formally identify hidden dynamic modes of the process driven by substantial external perturbations. A model-complexity index is introduced to quantify the degree of structural complication in the process and to indicate when additional boosting-based corrections to the forecast are required.

The method also incorporates a multiscale assessment of disturbance-intensity levels, an analysis of the dependence between the conditional mean and conditional variance, and clustering of the decision vectors of trees within the ensemble. This enables the detection of distinct patterns and modes without relying on auxiliary features. It is noted that the method can be extended to other classes of heteroscedastic process models that may be used in place of the EGARCH component.

An algorithm is proposed for identifying intervals of elevated volatility and heightened structural complexity, as well as for assessing their statistical significance using volatility-, entropy-, and econometrics-based metrics.

The practical application of the method is demonstrated using public air quality monitoring data for the city of Vinnytsia, specifically the PM1 indicator. The results show high predictive accuracy (coefficient of determination — 0.97) and make it possible to localize short-term stochastic disturbances and long-term structural anomalies that remain undetected by classical ARIMA, ARIMA-GARCH, Prophet, or multifactor intelligent models based on feature-engineering techniques such as those provided by the Python library tsfresh. A generalized analytical conclusion was generated automatically with the assistance of a Large Language Model. The study shows that the proposed structural-complexity index can reveal anomalies within the internal structure of the process that are not accompanied by increases in conditional variance and that reflect the emergence of new systematic influences or changes in the process’s dynamic mechanism.

The example provided confirms the effectiveness of the method for systems-analysis tasks and forms a basis for its further use in decision-support systems for assessing environmental conditions and other complex systems.

333 51

How to Cite

[1]
“Method of Structural Analysis of Heteroscedastic Time Series Using the Boosting EGARCH Model”, Вісник ВПІ, no. 6, pp. 139–148, Dec. 2025, doi: 10.31649/1997-9266-2025-183-6-139-148.

Author Biographies

V. Ye. Kopniak, Vinnytsia National Technical University

Post-Graduate Student of the Chair of System Analysis and Information Technologies

V. B. Mokin, Vinnytsia National Technical University

Dr. Sc. (Eng.), Professor, Head of the Chair of System Analysis and Information Technologies

References

В. Є. Копняк, В. Б. Мокін, С. О. Жуков, І. В. Варчук, і Т. В. Скринник, «Метод бустингу гетероскедастичних моделей для прогнозування концентрацій пилу Сахари в атмосферному повітрі України,» Наукові праці Наукові праці Вінницького національного технічного університету, вип. 2, Лип. 2024. [Електронний ресурс]. https://doi.org/10.31649/2307-5376-2024-2-28-38 .
J. D. Hamilton, “A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle,” Econometrica, vol. 57, no. 2, pp. 357-384, Mar. 1989. https://doi.org/10.2307/1912559 .
В. Б. Мокін, А. В. Лосенко, і А. Р. Ящолт, «Інформаційна технологія аналізу та прогнозування кількості нових випадків захворювань на коронавірус SARS-CoV-2 в Україні на основі моделі Prophet,» Вісник Вінницького політехнічного інституту, № 5, с. 71-83, 2020. https://doi.org/10.31649/1997-9266-2020-152-5-71-83 .
G. E. P. Box, G. M. Jenkins, G. C. Reinsel, and G. M. Ljung, Time Series Analysis: Forecasting and Control, 5th ed. Hoboken, NJ, USA: John Wiley & Sons, 2015, 712 p. ISBN: 978-1-118-67502-1.
В. Б. Мокін, і М. В. Дратований, Наука про дані: машинне навчання та інтелектуальний аналіз даних, електр. навч. посіб. Вінниця, Україна: ВНТУ, 2024, 258 с. [Електронний ресурс]. Режим доступу: https://docs.vntu.edu.ua/card.php?id=8163.
J. Brownlee, “How to Model Volatility with ARCH and GARCH for Time Series Forecasting in Python,” Machine Learning Mastery, Aug. 24, 2018. [Electronic resource]. Available: https://machinelearningmastery.com/develop-arch-and-garch-models-for-time-series-forecasting-in-python .
M. Christ, N. Braun, J. Neuffer, and A. W. Kempa-Liehr, “Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests (tsfresh – A Python package),” Neurocomputing, vol. 307, pp. 72-77, 2018. https://doi.org/10.1016/j.neucom.2018.03.067 .
A. Bhatnagar, P. Kassianik, C. Liu, et al., “Merlion: A Machine Learning Library for Time Series,” arXiv preprint, arXiv:2109.09265, 2021. https://doi.org/10.48550/arXiv.2109.09265 .
Q. P. Duy, et al., “Estimating and forecasting bitcoin daily prices using ARIMA-GARCH models,” Business Analyst Journal, vol. 45, no. 1, pp. 11-23, 2024. https://doi.org/10.1108/BAJ-05-2024-0027.
J. A. Mariño, M. E. Arrieta-Prieto, and S. A. Calderón, “Comparison between statistical models and machine learning for forecasting multivariate time series: An empirical approach,” Communications in Statistics: Case Studies, Data Analysis and Applications, vol. 11, no. 1, pp. 56-91, 2025. https://doi.org/10.1080/23737484.2025.2463905 .
J. Olbryś, and E. Majewska, “Regularity in stock market indices within turbulence periods: The sample entropy approach,” Entropy, vol. 24, no. 7, p. 921, 2022. https://doi.org/10.3390/e24070921 .
I. Chronopoulos, L. Giraitis, and G. Kapetanios, “Choosing between persistent and stationary volatility,” The Annals of Statistics, vol. 50, no. 6, pp. 3466-3483, Dec. 2022. https://doi.org/10.1214/22-AOS2236 .
G. Bardas, N. Kefalakis, and J. Soldatos, “Indicators of External Disruptions in Supply Chains: A Framework for Early Detection and Resilience Planning,” in 2025 21st International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT). Lucca, Italy, 2025, pp. 1064-1071. https://doi.org/10.1109/DCOSS-IoT65416.2025.00160 .
K. Shibano, R. Lin, and G. Mogi, “Volatility Reducing Effect by Introducing a Price Stabilization Agent on Cryptocurrencies Trading,” in Proceedings of the 2020 2nd International Conference on Blockchain Technology (ICBCT ’20). New York, NY, USA: ACM, 2020, pp. 85-89. https://doi.org/10.1145/3390566.3391679 .
V. Mokin, D. Shmundiak, and V. Kopniak, “Air Quality Monitoring from EcoCity,” Kaggle Dataset, May 2024. [Electronic resource]. Available: https://www.kaggle.com/datasets/vbmokin/air-quality-monitoring-from-ecocity .

Most read articles by the same author(s)

1 2 3 4 5 6 7 > >>