Masanobu Taniguchi (Waseda University) Joint work with Hiroshi Shiraishi (Keio University)
Date: 13 SEPTEMBER 2018 at 14:30
Event location: Aula III, 2nd floor, Department of Statistics, Via Belle Arti1 41, Bologna
Type: Statistics Seminars
Abstract:
Optimal portfolios are determined by the mean and variance of the portfolio return. Several authors proposed estimators of the optimal portfolios as the functions of the sample mean and the sample variance for independent returns of assets.
However, many empirical studies show that return processes are generally non-Gaussian and dependent.
Basak, Jagannathan and Sun (2002) showed the consistency of optimal portfolio estimators when the portfolio returns are stationary processes.
However, in the literature, there has been no study on the asymptotic efficiency of estimators for optimal portfolios. In this presentation, denoting the optimal portfolios by a function of the mean and covariance matrix of the return process concerned, we discuss the asymptotic efficiency of estimators when the returns are vector-valued non –Gaussian (locally) stationary process