| Annotation: | Methods. The application of the abstraction method allowed for the isolation of volatility characteristics, simplifying the analysis of complex financial data of the cryptocurrency market. Analysis with synthesis facilitated the identification of patterns and the integration of traditional and modern forecasting approaches, providing a comprehensive assessment of methods. Logical and historical approaches enabled evolutionary analysis, while classification methods based on general and specific analysis principles, combined with comparative and abstract-logical analysis, allowed for an objective evaluation of the developed models’ effectiveness and justified the feasibility of developing innovative solutions for optimizing trading strategies and minimizing risks. Results. The study conducted a comparative analysis of cryptocurrency market volatility prediction methods using traditional statistical approaches and modern machine learning algorithms. The results confirm the advantages of integrating classical methods with machine learning algorithms, which allow for more accurate risk assessment and optimization of trading strategies in the highly volatile cryptocurrency markets. The determined volatility can be used in conjunction with Reinforcement Learning (RL) to optimize trading strategies, allowing an agent to learn to make decisions in an environment to maximize cumulative reward. The use of RL in cryptocurrency trading is a promising direction but requires a cautious approach and thorough testing of strategies before their application in real trading.Novelty. The scientific novelty lies in a comprehensive approach to forecasting cryptocurrency market volatility, combining classical statistical methods with modern machine learning algorithms. The advantages of ensemble machine learning methods for analyzing cryptocurrency volatility have been established. The integration of Reinforcement Learning (RL) for optimizing trading strategies based on predicted volatility is proposed, representing a new approach to cryptocurrency trading automation. Practical value. The research results have practical significance for cryptocurrency market participants, including investors, traders, and financial analysts. The integration of machine learning methods with traditional statistical approaches opens new opportunities for developing effective trading strategies, contributing to increased profitability and stability in the cryptocurrency market. The research is also useful for developers of trading platforms and analytical tools, as it provides empirical data for improving prediction algorithms and market data analysis. |
| Literature: | - 1. Bouri, E., Saeed, E., Vinh Vo, X., and Roubaud, D. (2021). Quantile connectedness in the cryptocurrency market, Journal of International Financial Markets, Institutions and Money, Vol. 71, 101302. https://doi.org/10.1016/j.intfin.2021.101302.2.
- 2. Gupta, H., and Chaudhary. R. (2022)/ An Empirical Study of Volatility in Cryptocurrency Market. Journal of Risk and Financial Management. 15: 513. https://doi.org/10.3390/jrfm15110513
- 3. Time Series Analysis: Forecasting and Control, 5th Edition, by George E.P. Box, Gwilym M. Jenkins, Gregory C. Reinsel and Greta M. Ljung (2015). John Wiley and Sons Inc., Hoboken, New Jersey, pp. 712.
- 4. Kejia, Y., Yan, H., and Gupta, R. (2022). Are GARCH and DCC values of 10 cryptocurrencies affected by COVID‐19? Journal of Risk and Financial Manage- ment, 15(3), 113. https://doi.org/10.3390/jrfm15030113
- 5. Khedr, A. M., Arif, I., Pravija Raj, P. V., El- Bannany, M., Alhashmi, S. M., & Sreedharan, M. (2021). Cryptocurrency price prediction using traditional statistical and machine learning techniques: A survey. Intelligent Systems in Accounting, Finance and Management, 28(1), 3-34. https://doi.org/10.1002/isaf.1488
- 6. Bollerslev, Tim. Generalized autoregressive conditional heteroskedasticity, Journal of Econometrics, Volume 31, Issue 3, 1986, Pages 307-327, https://doi.org/10.1016/0304-4076(86)90063-1.
- 7. Henrique, B. M., Sobreiro, V. A., & Kimura, H. (2019). Literature review: Machine learning techniques applied to financial market prediction R. Expert Systems with Applications, 124, 226-251. https://doi.org/10.1016/j.eswa.2019.01.012
- 8. Umar Yahaya, H., Sunday Oyinloye, J., & Olorunfemi Adams, S. (2022). Modeling and Forecasting Cryptocurrency Returns and Volatility: An Application of GARCH Models. Universal Journal of Finance and Economics, 2(1), 71-90. Universal Journal of Finance and Economics, 2022, 2, 497 https://doi.org/10.31586/ujfe.2022.497
- 9. Lahmiri, S., & Bekiros, S. (2019). Cryptocurrency forecasting with deep learning chaotic neural networks. Chaos, Solitons & Fractals, 118, 35-40. https://doi.org/10.1016/J.CHAOS.2018.11.014
- 10. Nazareth, N., & Ramana Reddy, Y. V. (2023). Financial applications of machine learning: A literature review. Expert Systems with Applications, 219(January), 119640. https://doi.org/10.1016/j.eswa.2023.119640
|
Archive