پیش بینی نوسانات بازدهی ارزهای رمزنگاری شده با استفاده از روشهای مارکوف پنهان و گارچ – مارکوف
محورهای موضوعی : بورس اوراق بهادار
مریم باقرزاده سهرابی
1
,
حسین ممبینی
2
*
,
صفیه مهری نژاد
3
1 - گروه مدیریت مالی، واحد تهران مرکزی، دانشگاه آزاد اسلامی، تهران، ایران.
2 - گروه مدیریت مالی، واحد تهران مرکزی، دانشگاه آزاد اسلامی، تهران، ایران
3 - گروه مدیریت مالی، واحد تهران مرکزی، دانشگاه آزاد اسلامی، تهران، ایران
کلید واژه: روش مارکوف پنهان, روش گارچ, اتریوم, بیت کوین,
چکیده مقاله :
پیش‌بینی دقیق در بازار ارزهای دیجیتال نقش اساسی در مدیریت ریسک، بهینه سازی پرتفوی دارد و به سرمایه گذاران در تصمیم گیری آگاهانه تر، مدیریت ریسک و حداکثرسازی بازده کمک خواهد کرد . این پژوهش با توجه به اهمیت روز افزون دو رمز ارز بیت کوین و اتریوم، براساس اطلاعات روزانه، اقدام به پیش بینی بازدهی و نوسانات بازدهی آنها با روش مارکوف پنهان و مدل ترکیبی گارچ - مارکوف پنهان نموده است. براساس نتایج پیش بینی، دقت پیش بینی بازدهی اتریوم براساس روش مارکوف پنهان با شاخص DAP=65.38% و با شاخص MAPE=1.76% بوده است. استفاده از روش گارچ مارکوف منجر به بهبود 25 درصد در دقت پیش بینی روند شده است اما دقت پیش بینی قدرمطلق مقادیر در قیاس با مدل مارکوف پنهان کمی بدتر شده است. برای رمز ارز بیت کوین نیز نتایج پیش بینی حاکی از آن است که دقت مدل براساس شاخص DAP=76.92% و براساس شاخص MAPE=1.43% بوده است. پس از بکارگیری روش گارچ - مارکوف عملکرد مدل در پیش بینی روند 5درصد بهتر شده است اما شاخص قدرمطلق مقادیر 55 درصد شده است.
According to theoretical literature, accurate prediction in the digital currency market plays a fundamental role in risk management, portfolio optimization . It assists investors in making more informed decisions, managing risk, and maximizing returns. Based on the importance of the ever-growing cryptocurrencies Bitcoin and Ethereum, this research has predicted their performance and yield fluctuations using a combination of Hidden Markov Models and the Garch-Markov method, based on daily data. Based on the prediction results, the accuracy of Ethereum’s performance prediction using the Hidden Markov Model method with the DAP = 65.38%, and with the MAPE = 1.76%. Therefore, using this method has led to a 25% improvement in trend prediction accuracy. However, the absolute prediction accuracy for values has slightly worsened compared to the simple Markov model. For the cryptocurrency Bitcoin, the prediction results indicate that the model’s accuracy based on the DAP = 76.92%, and based on the MAPE = 1.43%. After employing the GARCH-Markov method, the model’s performance in predicting trends has improved by 5%, but the absolute value index has reached 55%.
1) Adebiyi, A. A., Adewumi, A.O., & Ayo, C. K. (2014). Comparison of ARIMA and artificial neural networks models for stock price prediction. Journal of Applied Mathematics, 2014, 1-7 2) Al Galib, A., Alam, M. and Rahman, R.M. (2014) Prediction of stock price based on hidden Markov model and nearest neighbour algorithm’, Int. J. Information and Decision Sciences, Vol. 6, No. 3, pp.262–292. 3) Antonello Maruotti, Antonio Punzo, Luca Bagnato, Hidden Markov and Semi-Markov Models with Multivariate Leptokurtic-Normal Components for Robust Modeling of Daily Returns Series, Journal of Financial Econometrics, Volume 17, Issue 1, Winter 2019, Pages 91–117, https://doi.org/10.1093/jjfinec/nby019 4) Atsalakis, G. S., & Valavanis, K. P. (2009a). Forecasting stock market short-term trends using a neuro-fuzzy based methodology. Expert Systems with Applications, 36(7), 10696-10707 . 5) Bildirici, M., Ersin, O. O. (2009). “Improving forecasts of GARCH family models with the artificial neural networks: An application to the daily returns in Istanbul stock exchange” Expert Systems with Application. 6) Cao, W., Zhu, W., & Demazeau, Y. (2019). Multi-Layer Coupled Hidden Markov Model for CrossMarket Behavior Analysis and Trend Foreca. 7) Cavalcante, R. C., Brasileiro, R. C., Souza, V. L.F., Nobrega, J. P., & Oliveira, A. L.I. (2016). Computational Intelligence and Financial Markets: A Survey and Future Directions. Expert Systems with Applications, 55, 194-21. 8) Fetroos, M, Miri, E and Ayoub Miri.(2020). Comparison of Portfolio Optimization for Investors at Different Levels of Investors' Risk Aversion in Tehran Stock Exchange with Meta-Heuristic Algorithms, Advances in Mathematical Finance and Applications. 1(15). https://doi.org/10.22034/amfa.2019.1870129.1235 9) Gupta, A., & Dhingra, B. (2012, March). Stock market prediction using hidden Markov models. In Engineering and Systems (SCES), 2012 Students Conference on (pp. 1-4). IEEE. 10) Hassan, M. R., & Nath, B. (2005, September). Stock market forecasting using hidden Markov model: a new approach. In Intelligent Systems Design and Applications, 2005. ISDA'05. Proceedings. 5th International Conference on (pp. 192-196). IEEE. 11) Hassan, M. R. (2009). A combination of hidden Markov model and fuzzy model for stock market forecasting. Neurocomputing, 72(16), 3439-3446. 12) JAROSLAV LAJOS,(2011)” Computer Modeling Using Hidden Markov Model Approach Applied to the financial ”Doctoraldissertation, Oklahoma State University,United states of America 13) Li, J., Pedrycz, W., Wang, X. et al. A Hidden Markov Model-based fuzzy modeling of multivariate time series. Soft Comput 27, 837–854 (2023). https://doi.org/10.1007/s00500-022-07623-6 14) Naderi. H, Ganbari, M, Jamshidi, B and Aash nademi. (2024). The improved Semi-parametric Markov switching models for predicting Stocks Prices, Advances in Mathematical Finance and Applications, https://doi.org/10.22034/amfa.2021.1923297.1565
15) Padmaja Dhenuvakonda, R. Amandan, N. Kumar,(2020, November), “Stock Price Prediction Using Artificial Neurl Networks “ ,Journal of Critical Reviews ,Vol 7, pp.846-850.
16) Ritesh Patel, Mariya Gubareva, Muhammad Zubair Chishti (2024) Assessing the connectedness between cryptocurrency environment attention index and green cryptos, energy cryptos, and green financial assets, Research in International Business and Finance, Volume 70,.https://doi.org/10.1016/j.ribaf.2024.102339.
17) Simran, Anil Kumar Sharma, Asymmetric impact of economic policy uncertainty on cryptocurrency market: Evidence from NARDL approach, The Journal of Economic Asymmetries, https://doi.org/10.1016/j.jeca.2023.e00298.
18) Shou, M.-H., Wang, Z.-X., Li, D.-D. and Zhou, Y.-T. (2021), "Forecasting the price trends of digital currency: a hybrid model integrating the stochastic index and grey Markov chain methods", Grey Systems: Theory and Application, Vol. 11 No. 1, pp. 22-45. https://doi.org/10.1108/GS-12-2019-0068
19) Tabar, S., Sharma, S., & Volkman, D. (2020). A new method for predicting stock market crashes using classification and artificial neural networks. International Journal of Business and Data Analytics, 1(3), 203-217.
20) Tkáč, M., & Verner, R. (2016). Artificial neural networks in business: Two decades of research. Applied Soft Computing, 38(1), 788-804.
21) Wang, S. (2020, February). The Prediction of Stock Index Movements Based on Machine Learning. In Proceedings of the 2020 12th International Conference on Computer and Automation Engineering (pp. 1-6).
22) Yan, D., Zhou, Qi, Wang, J., & Zhang, N. (2017). Bayesian regularisation neural network based on artificial intelligence optimisation. International Journal of Production Research, 55(8), 2266-2287