Automated Trading Bot Using Hidden Markov Models for Cryptocurrency Prediction

The development of financial technology and the increasing popularity of cryptocurrency trading have significantly driven the demand for efficient automated trading bots capable of accurate and rapid price prediction and trading decision-making. This study aims to develop an automated trading bot using Hidden Markov Models (HMM) for cryptocurrency price prediction to capitalize on market volatility effectively. The experimental results demonstrate that HMM achieves a high prediction accuracy of 80.08% and effectively mitigates risks associated with outdated data usage. Compared to the Decision Tree model, which has an accuracy of 70.78%, HMM also shows superior profitability by 80% in dynamic trading scenarios within the Bitcoin market.