时 间：2025年5月20 日（周二）13：40 -14：20
报告人：马成虎，复旦大学教授
地 点：普陀校区理科大楼A1514
主持人：周勇 华东师范大学教授
摘 要：
In this paper, we investigate whether Knightian uncertainty can be quantified and separated from the conventional risk when the underlying return process belongs to the class of set-valued location-scale (LS) models driven by independent and identically distributed Gaussian random seeds. We establish impossibility theorems demonstrating that no single uncertainty measure can be universally agreed among investors with general risk-and-uncertainty-averse preferences. Nevertheless, it is always possible to rank the degree of uncertainty between two return processes by examining their mean-average characteristic functions. Specifically, under a rectangular LS-range specification, we show that the uncertainty, characterized by the ranges of the LS coefficients, can be fully identified from the first, second, fourth, and sixth moments of the return process. We also correct the composite uncertainty measure proposed by Izhakian (2020). Using data on stock market indices in the U.S. (S&P 500) and China (CSI 300), we estimate the proposed uncertainty measures and confirm that uncertainty and risk are indeed distinct concepts and can be disentangled empirically.
报告人简介：
马成虎，加拿大多伦多大学经济学博士，复旦大学管理学院财务金融系教授、金融工程硕士项目学术主任。先后在山东大学数学系、加拿大McGi大学经济系、英国Essex大学财务金融管理系、厦门大学王亚南经济研究院、复旦大学管理学院任教;日本京都大学经济研究所访问教授、新加坡国立大学经济系资深访问学者。研究涉及资产定价理论、博弈论、决策论、风险测量与管理等多个相关领域。主持国家自然科学基金、加拿大SSHRC和FCAR、英国ESRC等国家级研究课题5项，出版《金融经济学原理》、《高级资产定价理论》及“Advanced Asset Pricing Theory”等专著三部。发表学术论文30余篇，研究成果具有国际影响力。论文发表在包括 Economic Theory, Journal of Economic Dynamics & Control, Journalof Mathematical Economics, Mathematical Finance, social choice and Welfare等国际著名经济学和金融学期刊。
