The integration of artificial intelligence in financial markets, particularly in intraday trading, is rapidly evolving. Zhuokai Chen’s study explores a novel approach by incorporating large language models (LLMs) into a trading framework for the Nasdaq 100. This innovative strategy merges a volatility breakout system with AI-driven sentiment analysis, utilizing the DeepSeek Chat API to enhance trade decisions based on macroeconomic events like FOMC meetings and CPI releases. Over a seven-year backtesting period, the AI-enhanced method achieved impressive results, including a cumulative return of nearly 1,000% and a Sharpe ratio of 2.11. Despite these promising outcomes, the study’s reliance on backtesting poses limitations, as real-world factors like slippage and order delays were not accounted for. Furthermore, the proprietary nature of the LLM filtering system raises concerns about reproducibility and the risk of output drift. For broader adoption, further validation, live testing, and increased transparency are necessary. The study highlights the potential of AI in refining trading strategies but suggests that additional enhancements are needed before full-scale implementation.
How Can AI Revolutionize Intraday Trading with Economic Indicators?
