财务与金融系学术讲座系列2021年第10讲

主 题：机器是否能理解人类的决策？剖析股票预测技巧

主讲人: Baozhong Yang（金融学副教授，佐治亚州立大学J. Mack Robinson商学院）

协调人: 李晨煦 任泓霖

时 间：2021-10-08 09:00

地 点：Zoom会议室

语 言：英文

地点： Zoom会议室

https://zoom.us/j/88150349912?pwd=TjUzWkRpa2dkNUdmaXJiZ2VJZEFkZz09

会议 ID：881 5034 9912

密码：412961

讲座摘要:

Human decisions are important but difficult to understand or predict. This paper uses machine learning models, which are adept at capturing nonlinear and complex relations, to analyze analysts' forecasting decisions and determine their skill. Machine-identified skilled analysts persistently outperform human expert-picked star analysts. Machines rely on nonlinear interactions of analyst characteristics, such as past skill and efforts, to identify analyst skill, in contrast with human experts, who lean more on relation-based information such as brokerage size. The puzzle of post-analyst revision drifts can be explained by our model in that such drifts are concentrated in machine-picked skilled analysts. Our approach also allows the formation of a "smart'' analyst consensus that aggregates the forecasts of machine-picked skilled analysts. Investment strategies based on revisions of machine-identified skilled analysts and the smart analyst consensus both generate significant abnormal returns. Overall, we propose an interpretable machine learning framework that can be used to analyze and evaluate human opinions in general settings such as online discussions, political forecasts, and macroeconomic outlooks.

主讲人简介:

Baozhong Yang is the H. Talmage Dobbs Jr Chair in Finance and Associate Professor of Finance at the J. Mack Robinson College of Business in Georgia State University. He is also the Director of the FinTech Lab at the Robinson College, one of the first such labs associated with a business school in the nation. He has co-organized the inaugural and second GSU-RFS FinTech Conferences, a leading FinTech conference that offers dual submission to the premier journal Review of Financial Studies. Professor Yang has also served as referees for the leading finance and economics journals and served on the Program Committees of prestigious conferences such as the Western Finance Association and the SFS Cavalcade meetings.

Professor Yang’s research interests span theoretical and empirical studies in FinTech, Investments, and Corporate Finance. His most recent research involves innovative applications of Machine Learning and AI to study economic questions in Capital Markets and Corporate Finance. Professor Yang’s research has been published in leading academic journals in finance, accounting, operations research, computer sciences, and mathematics, including the Journal of Finance, Journal of Financial Economics, Review of Financial Studies, Journal of Accounting Research, Management Science, Mathematics of Operations Research, IEEE Computer, and Advances in Mathematics. His research has been widely cited and recognized by prizes such as the Emerald Citations of Excellence, and Chicago Quantitative Alliance Annual Academic Competition Prize, and the Yihong Xia Best Paper Prize.

Professor Yang’s papers have been extensively presented at prestigious conferences, such as the National Bureau of Economic Research (NBER) Big Data, NBER Economics of AI, NBER Blockchain, NBER Law and Economics, American Finance Association, and Western Finance Association Meetings. He has been invited to present at leading universities, including Stanford University, UCLA, University of Maryland, University of Minnesota, and University of Toronto. Professor Yang’s research has been also widely covered by the media, including the NBER Digest, Bloomberg, Wall Street Journal, Financial Times, Forbes, The Guardian, CNBC, Chicago Booth Review, Columbia Law School Blog, Duke University FinTech Blog, and University of Oxford Business Law Blog.

Professor Yang received his Ph.D. in Finance from Stanford University and Ph.D. in Mathematics from the Massachusetts Institute of Technology. He has also been a Gold Medalist in the 33rd International Mathematical Olympiad while in high school.