شناسايي نقطه ورود بهينه معاملات در بازار ارزهاي ديجيتال با استفاده از يادگيري ماشين
محورهای موضوعی : پردازش چند رسانه ای، سیستمهای ارتباطی، سیستمهای هوشمند
ابوالفضل یاوری خلیل آباد
1
,
حسن اعمی بنده قرایی
2
,
سید محمدرضا هاشمی
3
1 - استاديار، گروه مهندسي کامپيوتر و فناوري اطلاعات، دانشگاه پيام نور، تهران، ايران
2 - استاديار، گروه مديريت اقتصاد و حسابداري، دانشگاه پيام نور، تهران، ايران
3 - استادیار، گروه مهندسي کامپيوتر، دانشگاه ملی مهارت، تهران، ايران
کلید واژه: بازار رمزارزها, پیش‌بینی بازار, شبکه عصبی مصنوعی, ماشین بردار پشتیبان, نزدیکترین همسایه, تحلیل تکنیکال,
چکیده مقاله :
در سال‌هاي اخير استفاده از مدل‌هاي يادگيري ماشين در حوزه‌هاي مختلفي مورد استفاده قرارگرفته است. از جمله اين حوزه‌ها مي‌توان به بازار رمزارزها اشاره کرد که بخش مهمي از معاملات روزانه در آن توسط ربات‌هاي معامله‌گر هوشمند انجام مي‌شود. هدف اصلي‌اين مقاله، يافتن بهترين نقطه ورود به يک معامله با کمک روش هاي يادگيري ماشيني است. در‌اين مقاله از انديکاتورهاي شاخص قدرت نسبي (RSI) و ميانگين متحرک ساده (SMA) براي استخراج چندين ويژگي روندي و تلاقي استفاده شده است. سپس با استفاده از‌اين ويژگي‌ها، مدل‌هاي يادگيري شبکه عصبي چندلايه، ماشين بردار پشتيبان و نزديکترين همسايه آموزش داده مي‌شوند. در نهايت، عملکرد هريک از مدل‌ها بر روي داده‌هاي بازار ارزهاي ديجيتال در بازه هفت ماه نخست سال 2023 براي چندين ارز ديجيتال ارزيابي شده است. نتايج نشان مي‌دهند که اولاً ويژگی هاي استخراج شده از کارايي مناسبي برخوردارند و ثانياً مدل نزديکترين همسايه، نسبت به مدل هاي ديگر، بيشترين سود را در اين بازه زماني کسب کرده است.
Introduction: In the domain of financial forecasting, machine learning (ML) models have garnered significant attention in recent years. One prominent application lies in the cryptocurrency market, where intelligent trading bots facilitate a substantial portion of daily transactions.
Method: This paper investigates the efficacy of ML-based methods for identifying optimal trade entry points. Specifically, we employ the Relative Strength Index (RSI) and Simple Moving Average (SMA) indicators to extract a set of trend and crossover features. Subsequently, these features are utilized to train multilayer neural network, support vector machine, and nearest neighbor learning models. The performance of each model is then evaluated using digital currency market data encompassing the first seven months of 2023 for a variety of cryptocurrencies.
Results: Our findings demonstrate that, firstly, the extracted features exhibit promising efficacy. Secondly, the nearest neighbor model achieved the highest profitability during the evaluation period compared to the other investigated models.
Discussion: In the research conducted in this field, technical indicators are often used directly in market forecasting but in the proposed method of this article, instead of directly using the values of the indicators as the input of the classification methods, trend behaviors and their intersections have been used. In the continuation of this research, it is possible to determine the best exit points from the transaction with the help of machine learning and the use of volume indicators in the learning process.
[1] E. Fama. “Efficient capital markets: A review of the theory”. The Journal of Finance, 25, 383–417, 1970
[2] J. Almeida, T. Cruz,”A systematic literature review of investor behavior in the cryptocurrency markets”, Journal of Behavioral and Experimental Finance, Volume 37, 2023
[3] V. Derbentsev, V. Babenko, K. Khrustalev, H. Obruch and S. Khrustalova, “Comparative performance of machine learning ensemble algorithms for forecasting cryptocurrency prices”. International Journal of Engineering, 34, 140–148, 2021
[4] C. Conrad, A. Custovic and E. Ghysels, “Long-and short-term cryptocurrency volatility components: A GARCH-MIDAS analysis”. Journal of Risk and Financial Management, 11(2), 23, 2018
[5] Y. Liu and A. Tsyvinski, “Risks and returns of cryptocurrency”. The Review of Financial Studies, 34(6), 2689–2727, 2021
[6] R. C. Phillips and D. Gorse, “Predicting cryptocurrency price bubbles using social media data and epidemic modelling”. In 2017 IEEE Symposium Series on Computational Intelligence (SSCI) (pp. 1-7). IEEE, 2017
[7] F. Valencia, A. Gomez-Espinosa and B. Valdes-Aguirre, “Price movement prediction of cryptocurrencies using sentiment analysis and machine learning”. Entropy, 21(6), 589, 2019
[8] A. Yavari, H. Hassanpour, B. Rahimpour and M. Mahdavi, “Event prediction in social network through Twitter messages analysis. Social Network Analysis and Mining”. 12. 10.1007/s13278-022-00911-x. 2022
[9] E. Bouri, C. K. Lau, B. Lucey and D. Roubaud, “Trading volume and the predictability of return and volatility in the cryptocurrency market”. Finance Research Letters, 29, 340–346, 2019
[10] W. Zhengyang, L. Xingzhou, R. Jinjin and K. Jiaqing, “Prediction of cryptocurrency price dynamics with multiple machine learning techniques”. In Proceedings of the 2019 4th International Conference on Machine Learning Technologiesو (pp. 15–19), 2019
[11] Hajighorbani, Hashemi, Broumandnia, faridpour, “a review of some semi-supervised learning methods”, Journal of Knowledge-Based Engineering and Information”, 2(4), 2013
[12] S. Ansaripour, T. J. Gandomani, “Comparison of the classification methods in software development effort estimation”, Intelligent Multimedia Processing and Communication Systems(IMPCS), 3(2), 11-18, 2022
[13] T. E. Koker and D. Koutmos, “Cryptocurrency trading using machine learning”. Journal of Risk and Financial Management, 13(8), 178, 2020
[14] G. Attanasio, L. Cagliero, P. Garza and E. Baralis, “Quantitative cryptocurrency trading: Exploring the use of machine learning techniques”. In In Proceedings of the 5th Workshop on Data Science for Macro-Modeling with Financial and Economic Datasetsو (pp. 1–6). 2019.
[15] E. Akyildirim, A. Goncu and A. Sensoy, “Prediction of cryptocurrency returns using machine learning”. Annals of Operations Research, 297(1), 3–36, 2021
[16] Z. Dongna, H. Xihui, L. Chi Keung and X. Bing, “Implications of cryptocurrency energy usage on climate change”, Technological Forecasting and Social Change”, Volume 187, 2023,
[17] M. Kumbure, C. Lohrmann, P. Luukka and J. Porras, “Machine learning techniques and data for stock market forecasting: A literature review”, Expert Systems with Applications”, Volume 197, 2022
