讲座摘要

Using an augmented covariance matrix as a bridge, we prove that the covariance matrix of asset returns can be isomorphic to a block diagonalized matrix, which enables market portfolio as the first principal component by a quasi-PCA. This framework provides a bridge to connect the empirical and theoretical asset pricing models, and provides a novel explanation for the failure of estimated market betas. By the block diagonalized matrix, we propose two necessary conditions to be new risk factors and an approach for testing. We evaluate 21 representatives and find that most of them do not perform well. Although some candidates provide information to asset returns, there exhibit significant substitutions to the effect of market portfolio.

主讲人简介

王志诚，北京雁栖湖应用数学研究院数字经济实验室研究员，北京大学光华管理学院金融系副教授。最早在国内开设金融计量和信用风险管理等课程，主要研究领域为风险管理、金融计量模型、公司财务和数字经济。学术研究成果发表在Journal of Econometrics、Journal of Bank & Finance、《管理世界》、《经济研究》、《金融研究》等国内外顶级期刊。研究成果曾获中国国际金融年会优秀论文奖，2019年在国际计量经济学顶级期刊《Journal of Econometrics》发表改进VaR的MMVaR方法，关于区块链和数据经济方面的观点发表于国内多家主流媒体。