一、主题：What Firm Characteristics Drive US Stock Returns?

二、主讲人：Yufeng Han，北卡大学夏洛特分校贝克商学院，金融学终身副教授。2003年获得圣路易斯华盛顿大学金融博士学位，硕士就读于纽约州立大学布法罗分校，本科就读于清华大学。研究领域包括资产定价、经济计量学、贝叶斯经济计量学、投资管理、国际金融、房地产。研究成果曾在Journal of Financial Economics、Journal of Banking and Finance、Journal of Financial and Quantitative Analysis、《金融研究》等国内外高水平金融学刊物发表。曾获得科州大学丹佛分校授予的“杰出科研创新奖”，是多个国际著名期刊的匿名审稿人。在北卡大学夏洛特分校任教之前，曾在杜兰大学和科州大学丹佛分校任教。

三、时间：2018年7月4日，中午12:30-1:30

四、地点：中央财经大学主教913会议室

五、主持人：黄志刚，中央财经大学金融学院副教授

Abstract: Considering a comprehensive set of 94 firm characteristics, Green, Hand, and Zhang (2017) and that the predictive ability of firm characteristics for US stock returns declines substantially after 2003 and conclude that only two characteristics affect cross-sectional value-weighted expected returns since that time. Instead of relying on conventional ordinary or weighted least squares to estimate high-dimensional linear regressions, which is susceptible to overfitting, we apply a robust forecast combination approach to the cross section of returns. Using machine learning tools to pool forecasts, we find that most of the firm characteristics matter over time -- and approximately 30 matter on average at each point in time -- for cross-sectional expected returns both before and after 2003. Our combination approach provides informative and economically significant forecasts of cross-sectional returns before and after 2003.