عليرضا غريب شاهيان

The cryptocurrency market, characterized by extreme volatility, nonlinear dynamics, an‎d rapid structural shifts, requires the development of data-driven trading strategies that can adapt to highly dynamic conditions. This research aims to design, implement, an‎d eva‎luate an innovative trading strategy titled Volatility-Adaptive Momentum Network (VAM-Net), developed to deliver stable an‎d reliable performance in turbulent markets. The model is built upon three theoretical components: momentum extraction, market volatility-regime detection, an‎d a machine-learning-based decision layer, addressing the limitations of single-dimension trading methods in highly volatile environments. A systematic literature review an‎d research-gap analysis were conducted to identify the role of traditional, technical, statistical, meta-heuristic, an‎d machine-learning approaches in algorithmic trading. The target market was determined based on expert consultation to include Bitcoin, Ethereum, Binance Coin, Cardano, an‎d Ripple. A set of benchmark an‎d comparison models was selec‎ted, including momentum, mean reversion, moving average, gradient boosting, logistic regression, an‎d buy-an‎d-hold strategies. Furthermore, a comprehensive group of eva‎luation metrics across profitability, risk, an‎d performance-quality categories was defined an‎d computed using historical data from a reliable source. After implementing the models, all strategies were examined within an out-of-sample testing period. The results show that the proposed VAM-Net model outperforms competing strategies in most cryptocurrencies, particularly in terms of cumulative return, risk-adjusted ratios, signal stability, an‎d drawdown reduction. The Best–Worst Method (BWM) was applied for metric weighting based on expert judgment, an‎d final ranking of strategies was carried out using the TOPSIS method. The findings highlight that integrating trend information, volatility structure, an‎d data-driven learning can significantly enhance accuracy, profitability, an‎d stability in algorithmic trading within cryptocurrency markets.

معاملات الگوريتمي , استراتژي هاي معاملاتي , رمزارز , تصميم‌گيري چندمعياره

Algorithmic Trading , strategy trading , Cryptocurrency , Multi-criteria decision-making

Alireza Gharibshahian

Rouzbeh Ghousi