Dynamic Analysis of the Impact of Monetary Policy Changes on Derivatives Market Pricing Behavior from a Behavioral Finance Perspective

Abstract

This study examines the dynamic effects of monetary policy changes on derivatives pricing behavior, emphasizing applications in financial risk management for industrial commodities. High-frequency policy sentiment is extracted from textual announcements using BERT-based natural language processing, capturing market expectations, confidence, and uncertainty. Multi-source investor sentiment, including trading volume, option volatility, and social media indicators, is fused and incorporated into a time-varying parameter VAR model to analyze nonlinear pricing responses under heterogeneous emotional states. Market participant heterogeneity is further explored using HDBSCAN clustering, while Hawkes processes decompose trading intensity into baseline, sentiment-driven, and feedback-amplified components. The analysis reveals significant overreaction in derivative prices and heterogeneous contributions of different participant clusters following policy shocks. The framework is particularly suitable for integration with real-time high-frequency financial data platforms, wireless monitoring infrastructures, and edge-computing analytics, enabling timely risk assessment and adaptive decision-making for commodity and derivatives markets in an industrial management context.