From Market Volatility to Predictive Insight: An Adaptive Transformer-RL Framework for Sentiment-Driven Financial Time Series Forecasting

Financial time series prediction remains a significant challenge, driven by market volatility, nonlinear dynamic characteristics, and the complex interplay between quantitative indicators and investor sentiment. Traditional time series models (e.g., ARIMA, GARCH) struggle to capture the nuanced sentiment in textual data, while static deep learning integration methods fail to adapt to market regime transitions (bull markets, bear markets, consolidation). This study proposes a hybrid framework that integrates investor forum sentiment analysis with adaptive deep reinforcement learning (DRL) for dynamic model integration. By constructing a domain-specific financial sentiment dictionary (containing 16,673 entries) based on sentiment analysis approach and word embedding technique, we achieve 97.35% accuracy in forum title classification tasks. Historical price data and investor forum sentimental information are then fed into three Transformer variants (single-layer, multi-layer, bidirectional) and a support vector regressor (SVR) for predictions, with a Deep Q-Network (DQN) agent dynamically fusing the prediction results. Comprehensive experiments are conducted on diverse financial datasets, including China Unicom, CSI 100 Index, corn, and Amazon (AMZN). Experimental results demonstrate that our proposed approach, combining textual sentiment with adaptive DRL integration, significantly enhances prediction robustness in volatile markets, achieving the lowest RMSEs across diverse commodities. It overcomes the limitations of static methods and achieves cross-commodity generalization, outperforming both benchmark and state-of-the-art models.