Speaker: Kunpeng Li, professor and Dean, International School of Economics and Management, Capital University of Economics and Business

Date: December 1, 2023

Time: 9：30-10：30

Location: Tencent Meeting: 830-687-211

Sponsor: Zhongtai Securities Institute for Financial Studies, Shandong University

Abstract:

The partial least square (PLS) method gains growing popularity in empirical asset pricing literature.Compared with the traditional dimensional reduction method such as the principal components analysis (PCA)the PLS uses the predicting target to supervise the direction of dimension reduction. The literature has longbelieved that the PLS can consistently and efficiently recover the relevant factors. In this paper, we show that thisunderstanding is incorrect. The consistency of the PLS relies critically on what we call two-way orthogonalconditions which generally do not hold due to the fact that the relevant factor can be identified in the model. Dueto failure of the PLS method, we therefore propose a new method, that is called supervised expectation.maximization method (SEM), to recover the relevant factors. In the SEM, zero restrictions are exploited tosupervise the direction of dimension reduction. We establish the asymptotic properties of the SEM estimator. Werun simulations to investigate the performance of the SEM as well as the PCA and the PLS. The simulationresults present the bad performance of the PLS and the PCA, indicating that these two methods cannot extract theright direction even with the guidance of the predicting target. Our SEM method. as a comparison, works verywell in the same scenario. We apply our method to the application of the extra return of bonds and find betterforecasting performance.

For more information, please visit:

http://mathfinance.sdu.edu.cn/info/1273/7345.htm