vestreaВестник Российского экономического университета имени Г. В. ПлехановаVestnik of the Plekhanov Russian University of Economics2413-28292587-9251Plekhanov Russian University of Economics10.21686/2413-2829-2025-5-62-74vestrea-2460Research ArticleУПРАВЛЕНИЕ ИННОВАЦИЯМИINNOVATION MANAGEMENTНейронные сети для прогнозирования переходов рыночных режимов: эмпирическое исследование на рынке нефтиNeural Nets to Forecast Switches of Market Conditions: Empiric Research on Crude Oil MarketМанаховаИ. В.ManakhovaI. V.

Ирина Викторовна Манахова, доктор экономических наук, профессор,профессор кафедры

кафедра политической экономии

119991; Ленинские горы, д. 1; Москва

Irina V. Manakhova, Doctor of Economics, Professor, Professor of the Department

Department of Political Economy

119991; 1 Leninskie Gory; Moscow

manakhovaiv@mail.ruМатыцынА. В.MatytsynA. V.

Александр Владимирович Матыцын, соискатель ученой степени кандидата экономических наук

119991; Ленинские горы, д. 1; Москва

Alexander V. Matytsyn, Candidate of Sciences Degree in Economics

119991; 1 Leninskie Gory; Moscow

avmatytsyn@gmail.comЧередниченкоЛ. Г.CherednichenkoL. G.

Лариса Геннадиевна Чередниченко, доктор экономических наук, профессор, профессор кафедры

кафедра экономической теории

109992; Стремянный пер., д. 36; Москва

Larisa G. Cherednichenko, Doctor of Economics, Professor, Professor of the Department

Department of Economic Theory

109992; 36 Stremyanny Lane; Moscow

cherednlarisa@yandex.ruМосковский государственный университет имени М. В. ЛомоносоваРоссияLomonosov Moscow State UniversityRussian FederationРоссийский экономический университет имени Г. В. ПлехановаРоссияPlekhanov Russian University of EconomicsRussian Federation202530092025056274Copyright © Манахова И.В., Матыцын А.В., Чередниченко Л.Г., 20252025Манахова И.В., Матыцын А.В., Чередниченко Л.Г.Manakhova I.V., Matytsyn A.V., Cherednichenko L.G.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://vest.rea.ru/jour/article/view/2460

   В статье рассматривается задача автоматического прогнозирования переходов между различными рыночными режимами на рынке нефти с помощью нейросетевых моделей. Для анализа использованы данные по ценам нефти марки Brent за 2000–2025 гг. Классификация периодов рынка выполняется на основе скользящих статистик – волатильности, асимметрии и эксцесса, что позволяет автоматически выделять стабильные, волатильные и кризисные фазы. Сравнивается эффективность трех типов рекуррентных нейронных сетей: Simple RNN, LSTM и GRU. Выявлено, что более сложные архитектуры (LSTM, GRU) существенно превосходят базовую RNN по точности и полноте обнаружения событий смены режима. Авторы подчеркивают значение инженерии rolling-признаков и демонстрируют, что такой подход обеспечивает устойчивость и адаптивность моделей к изменчивости рынка. Результаты исследования показывают перспективность глубоких нейронных сетей для задач мониторинга и раннего предупреждения рыночных событий. В заключение обсуждаются ограничения подхода и направления дальнейших исследований, включая интеграцию внешних данных и развитие методов объяснимого искусственного интеллекта.

   The article studies the goal of automatic forecasting switches of different market conditions on oil market with the help of neural net models. For the analysis the authors used data on oil (Brent) prices in 2000-2025. Classification of market periods was done on the basis of fluctuating statistics – volatility, asymmetry and excess, which provides an opportunity to identify automatically stable, volatile and crisis phases. The efficiency of three types of competitive neural nets is compared: Simple RNN, LSTM and GRU. It was found that more complicated architecture (LSTM, GRU) surpasses the basic one (RNN) in accuracy and full identification of events of condition switch. The authors highlighted the importance of rolling-sign engineering and showed that this approach provides sustainability and adaptability of models to market changes. Research findings demonstrate promising nature of deep neural nets for monitoring and early warning of market events. Finally, restrictions of the approach were discussed, as well as trends of further investigations, including integration of external data and development of methods of explainable AI.

искусственный интеллектглубокое обучениегенеративно-состязательные сетимодель рекуррентных нейронных сетей (RNN)модель долгой краткосрочной памяти (LSTM)модель управляемых рекуррентных блоков (GRU)AIdeep traininggenerative-competitive netsmodel of recurrent neural nets (RNN)model of long short- term memory (LSTM)model of guided recurrent units (GRU)ReferencesЗямалов В. Е. Использование многорежимных моделей для моделирования динамики финансовых временных рядов // Экономическое развитие России. – 2022. – № 5. – С. 13–19.Zyamalov V. E. Ispolzovanie mnogorezhimnyh modeley dlya modelirovaniya dinamiki finansovyh vremennyh ryadov [The Use of Multi-Range Models for Modeling Dynamics of Finance Time Series]. Ekonomicheskoe razvitie Rossii [Economic Development of Russia], 2022, No. 5, pp. 13–19. (In Russ.).Латыпова Р. Р. Прогнозирование макроэкономических показателей на основе рекуррентной нейронной сети // Известия СПбГЭУ. – 2025. – № 1 (151). – C. 104–108.Latypova R. R. Prognozirovanie makroekonomicheskih pokazateley na osnove rekurrentnoy neyronnoy seti [Forecasting Macro-Economic Indicators Based on Recurrent Neuronal Net]. Izvestiya SPbGEU [Bulletin of SPbGEU], 2025, No. 1 (151), pp. 104–108. (In Russ.).Суворов В. А. Нейросетевое прогнозирование дебита нефти на этапе выработки запасов // Инновационная наука. – 2024. – № 4-1. – C. 65–73.Suvorov V. A. Neyrosetevoe prognozirovanie debita nefti na etape vyrabotki zapasov [Neural Forecast of Oil Yield at the Stage of Reserve Output]. Innovatsionnaya nauka [Innovation Science], 2024, No. 4-1, pp. 65–73. (In Russ.).Alruqimi M., Di Persio L. Enhancing Multi-Step Brent Oil Price Forecasting with Ensemble Multi-Scenario Bi-GRU Networks // arXiv:2407.11267v1. – 2024. – July 15.Alruqimi M., Di Persio L. Enhancing Multi-Step Brent Oil Price Forecasting with Ensemble Multi-Scenario Bi-GRU Networks. arXiv:2407.11267v1, 2024, July 15.An S., Gao X., An F., Wu T. Early Warning of Regime Switching in a Complex Financial System from a Spillover Network Dynamic Perspective // iScience. – 2025. – Vol. 28. – P. 111924.An S., Gao X., An F., Wu T. Early Warning of Regime Switching in a Complex Financial System from a Spillover Network Dynamic Perspective. iScience, 2025, Vol. 28, p. 111924.Awijen H., Ben Ameur H., Ftiti Z., Louhichi W. Forecasting Oil Price in Times of Crisis: a New Evidence from Machine Learning Versus Deep Learning Models // Annals of Operations Research. – 2023. – Vol. 345. – P. 979–1002.Awijen H., Ben Ameur H., Ftiti Z., Louhichi W. Forecasting Oil Price in Times of Crisis: a New Evidence from Machine Learning Versus Deep Learning Models. Annals of Operations Research, 2023, Vol. 345, pp. 979–1002.Chollet F. Deep Learning with Python. – 2nd ed. – Shelter Island : Manning Publications, 2021.Chollet F. Deep Learning with Python. 2nd ed. Shelter Island, Manning Publications, 2021.Foroutan P., Lahmiri S. Deep Learning Systems for Forecasting the Prices of Crude Oil and Precious Metals // Financial Innovation. – 2024. – Vol. 10. – N 1. – P. 1–40.Foroutan P., Lahmiri S. Deep Learning Systems for Forecasting the Prices of Crude Oil and Precious Metals. Financial Innovation, 2024, Vol. 10, No. 1, pp. 1–40.Fu X. Oil Price Forecasting Model Based on GARCH-LSTM Model // Frontiers in Business, Economics and Management. – 2023. – Vol. 10. – P. 28–31.Fu X. Oil Price Forecasting Model Based on GARCH-LSTM Model. Frontiers in Business, Economics and Management, 2023, Vol. 10, pp. 28–31.Li Q., Guo L., Wang J., Li X., Bai X. Oil Price Forecasting: a Hybrid GRU Neural Network Based On Variational Mode Decomposition // Energy Reports. – 2023. – Vol. 9. – P. 3185–3197.Li Q., Guo L., Wang J., Li X., Bai X. Oil Price Forecasting: a Hybrid GRU Neural Network Based On Variational Mode Decomposition. Energy Reports, 2023, Vol. 9, pp. 3185–3197.Lin S., Wang Y., Wei H., Wang X., Wang Z. Hybrid Method for Oil Price Prediction Based on Feature Selection and XGBOOST-LSTM // Energies. – 2025. – Vol. 18. – P. 2246.Lin S., Wang Y., Wei H., Wang X., Wang Z. Hybrid Method for Oil Price Prediction Based on Feature Selection and XGBOOST-LSTM. Energies, 2025, Vol. 18, p. 2246.Ly R., Traore F., Dia K. Forecasting Commodity Prices Using Long Short-Term Memory Neural Networks // arXiv:2101.03087v1. – 2021. – January 11.Ly R., Traore F., Dia K. Forecasting Commodity Prices Using Long Short-Term Memory Neural Networks. arXiv:2101.03087v1, 2021, January 11.Ma F., Wahab M. I. M., Huang D., Xu W. Forecasting the Realized Volatility of the Oil Futures Market: a Regime Switching Approach // Energy Economics. – 2017. – Vol. 67. – DOI: 10.1016/j.eneco.2017.08.004.Ma F., Wahab M. I. M., Huang D., Xu W. Forecasting the Realized Volatility of the Oil Futures Market: a Regime Switching Approach. Energy Economics, 2017, Vol. 67. DOI: 10.1016/j.eneco.2017.08.004.Mari D., Mari M. Deep Learning Based Regime-Switching Models of Energy Commodity Prices // Energy Systems. – 2022. – Vol. 13. – P. 359–376.Mari D., Mari M. Deep Learning Based Regime-Switching Models of Energy Commodity Prices. Energy Systems, 2022, Vol. 13, pp. 359–376.Ribeiro M., Singh S., Guestrin C. "Why Should I Trust You?": Explaining the Predictions of any Classifier // Proceedings of the 22<sup>nd</sup> ACM SIGKDD International Conference on Knowledge Discovery and Data Mining – KDD '16. – ACM Press, 2016. – P. 1135–1144.Ribeiro M., Singh S., Guestrin C. "Why Should I Trust You?": Explaining the Predictions of any Classifier. Proceedings of the 22<sup>nd</sup> ACM SIGKDD International Conference on Knowledge Discovery and Data Mining – KDD '16. ACM Press, 2016, pp. 1135–1144.Shahbazbagian V., Hosseininesaz H., Shafie-khah M., Elmusrati M. Forecasting Crude Oil Prices Using a Hybrid Model Combining Long Short-Term Memory Neural Networks and Markov Switching Model // International Conference on Future Energy Solutions (FES). – Vaasa, United States: University of Vaasa, 2023. – P. 1–6.Shahbazbagian V., Hosseininesaz H., Shafie-khah M., Elmusrati M. Forecasting Crude Oil Prices Using a Hybrid Model Combining Long Short-Term Memory Neural Networks and Markov Switching Model. International Conference on Future Energy Solutions (FES). Vaasa, United States, University of Vaasa, 2023, pp. 1–6.Tan Z., Wu Y. On Regime Switching Models // Mathematics. – 2025. – Vol. 13. – P. 1128.Tan Z., Wu Y. On Regime Switching Models. Mathematics, 2025, Vol. 13, p. 1128.Yang K., Hu N., Tian F. Forecasting Crude Oil Volatility Using the Deep Learning-Based Hybrid Models with Common Factors // Journal of Futures Markets. – 2024. – Vol. 44. – P. 1429–1446.Yang K., Hu N., Tian F. Forecasting Crude Oil Volatility Using the Deep Learning-Based Hybrid Models with Common Factors. Journal of Futures Markets, 2024, Vol. 44, pp. 1429–1446.Zhang K., Hong M. Forecasting Crude Oil Price Using LSTM Neural Networks // Data Science in Finance and Economics. – 2022. – Vol. 2. – P. 163–180.Zhang K., Hong M. Forecasting Crude Oil Price Using LSTM Neural Networks. Data Science in Finance and Economics, 2022, Vol. 2, pp. 163–180.Zhao Y., Hu B., Wang S. Prediction of Brent Crude Oil Price Based on LSTM Model under the Background of Low-Carbon Transition // arXiv:2409.12376v1. – 2024. – September 19.Zhao Y., Hu B., Wang S. Prediction of Brent Crude Oil Price Based on LSTM Model under the Background of Low-Carbon Transition. arXiv:2409.12376v1, 2024, September 19.

The authors declare that there are no conflicts of interest present.