Document Type : Research Paper
Authors
1 Department of Information Technology Management, Faculty of Management and Economics, Science and Research Branch, Islamic Azad University, Tehran, Iran.
2 Department of Information Technology Management, Faculty of Management and Economics, Science and Research Branch, Islamic Azad University, Tehran, Iran. Corresponding Author: mahmood_alborzi@yahoo.com
3 Department of Tourism Management, Faculty of Management and Accounting, Allameh Tabataba'i University, Tehran, Iran.
Abstract
Forecasting cryptocurrency market trends remains a significant challenge due to its fundamental differences from traditional currencies. This complexity arises from the interplay between conventional financial indicators, advancements in information technology, and government macroeconomic policies influencing market acceptance. This study introduces a novel decision support framework that, rather than analyzing individual cryptocurrencies, focuses on the overall acceptance of the cryptocurrency market. The proposed approach enables a more precise and realistic assessment of market trends, facilitating the generation of buy and sell guidance tables for any specified time interval. To achieve this, maximum likelihood estimation and Bayesian belief networks are employed, allowing for a comparative analysis of these methodologies. Additionally, a high-edge-strength Bayesian belief network is constructed from the generated networks to enhance prudent trading decisions. The method is validated using 1155 weekly and 484 daily time points across 21 cryptocurrencies with the highest market capitalization, covering two periods: the last quarter of 2024 and March–May 2025. The findings demonstrate that the proposed framework, with its high precision, accuracy, recall, and model robustness, supports buy and sell decisions with an average accuracy of at least 78% on a daily basis and 64.5% on a weekly basis. This approach offers a valuable tool for navigating the dynamic and uncertain nature of the cryptocurrency market.
Keywords
- conditional probabilities
- cryptocurrency
- maximum likelihood estimation
- Bayesian belief network
- decision support
Main Subjects
References [In Persian]
- Habibirad, A., & Panahi, A. (2021). Explaining the Relationship Between Bitcoin Price in Business Financial Transactions and Search Volume in Order to Identify its Behavioral Pattern: A Comparative Study Between Countries. Business Intelligence Management Studies, 10(37), 347-372. doi: 10.22054/ims.2021.61455.1982 [In Persian]
- Hosseingholizadeh, R., Alborzi, M., Toloie Eshlaghy, A., & Zargham Boroujeni, H. (2023). Iranian Tourism Stakeholders’ Opinion Dynamics in the Face of Chinese Tourists. Tourism Management Studies, 17(60), 47 -77. doi: 10.22054/tms.2022.69532.2746 [In Persian]
)