Future Price Monitoring by Using Large Language Model (LLM)

Futures price monitoring is important for identifying significant changes in financial markets. Traditional monitoring systems mainly use historical numerical data and statistical methods, which may not provide clear explanations for unusual price movements. This paper proposes a conceptual approach to futures price monitoring using a Large Language Model (LLM) and Retrieval-Augmented Generation (RAG). The proposed approach retrieves relevant financial information, such as market reports, financial news, and related documents, to provide contextual analysis of price movements. The LLM uses the retrieved information to generate understandable explanations and alerts that can help users interpret significant changes in futures prices. The approach aims to improve the interpretability of price monitoring by connecting market information with relevant financial knowledge. Important challenges include data quality, retrieval accuracy, processing latency, hallucinations, and computational cost. The proposed work presents a conceptual framework and does not claim experimentally validated performance results.