Seminar Calendar, Information and Innovation Management
Robust Minimum Variance Portfolio for a Large Universe of Assets
22 Feb 2023 | 10:30 a.m. — 12:00 noon
KK315
SPEAKER
Dr. Michael Fan
Assistant Professor
Department of Economics
The Chinese University of Hong Kong
ABSTRACT
This paper proposes a robust portfolio in a large-dimensional assets universe. We deal with the outliers or heavy tails in the return data using the approximate factor model with unobserved factors, which is the predominant workhorse for asset pricing (Fan et al., 2013). Specifically, our procedure employs a robust PCA method for estimating the factor model and a shrinkage estimation for the random errors covariance matrix. We develop the theoretical results of the robust portfolio, which accommodates both stationary and nonstationary data. Numerical and empirical results show the superior performance of the new portfolio.
