内容简介:

Weak factors—those to which test assets have limited exposure—pose a significant challenge to empirical asset pricing, and have been widely studied in the literature. Meanwhile, despite its relevance in empirical applications, the selection of test assets has received comparatively less systematic study. We introduce supervised-PCA (SPCA), a novel methodology to estimate factor risk premia that addresses both of these problems, by bridging these seemingly unrelated strands of the literature. SPCA iterates supervised selection of test assets, principal-component estimation, and factor projection. By optimally selecting test assets, it enables risk premia estimation and factor model diagnosis even when weak factors are present and not all factors are observed. We establish SPCA’s asymptotic properties and showcase its empirical applications.

演讲人简介：

Dake Zhang is currently a PhD student at the University of Chicago, Booth School of Business. Previously, he obtained a MS in Statistics from University of Chicago and a BS in Mathematics from Tsinghua University. His research interest includes developing statistical and machine learning methodologies, with a focus on their application in finance. Specifically, he works on the theories of factor models and principal component analysis, while exploring their application in asset pricing.