金融与财务学系列讲座之334期

Time：14:00-15:30, Tuesday, April 28, 2026

Location: Room 603, Starr Building（史带楼603室）

Host： Dr. Zhe Geng （耿哲 博士）

Department of Finance, FDSM

Topic: LLM-Based Expectations and Fed Communication

Speaker: Prof. Bin Li（李斌 教授）

Wuhan University

Abstract: We introduce a new approach to measuring monetary policy surprises in FOMC press conferences. Our method uses LLMs to generate counterfactual press-conference responses by the Federal Re- serve Chair and measures policy surprises as deviations between realized and simulated answers. Using high-frequency intraday data from 2011 to 2025, we show that equity prices respond strongly to these surprises: hawkish surprises lead to significant declines in stock prices, while dovish sur- prises increase prices. Importantly, markets respond to the unexpected component of communi- cation rather than to the level of expressed policy stance. Heterogeneity tests show that market responses to policy surprises operate through expectation formation and belief updating. Reactions are strongest when the Chair’s answers directly inform the future policy path, namely for forward- looking, policy-related, or statement-linked questions, and are larger, though slower to emerge, during periods of elevated market uncertainty. Responses are also stronger following dovish FOMC statements, when prior expectations are less anchored. Finally, omitting macroeconomic context or the policy statement from the LLM prompt eliminates the return predictability of policy sur- prises. Our findings demonstrate that LLMs provide a powerful tool for identifying economically meaningful surprises in real-time policy communication.

Bio: 李斌，现为武汉大学经济与管理学院金融系教授，获教育部青年**学者称号，研究领域为金融科技、金融机器学习和量化投资管理等。他具备金融+科技的跨学科背景与研究能力，研究成果发表于《Journal of Financial Economics》《Journal of Accounting Research》《管理科学学报》《金融研究》等经管类期刊，以及《Artificial Intelligence》《Journal of Machine Learning Research》《ICML》等计算机类期刊会议；主持国家自然科学基金、社科基金等项目，两项自科基金后评估为特优。他是特许金融分析师(CFA)持证人，担任武汉大学教育发展基金会理事和投资咨询委员会副主任，兼任教育部高等学校金融类专业教学指导委员会委员等。