Lunch Seminar Mathematics and Statistics - 2021/2022
Interviene: Anlan Wang, VSB – Technical University of Ostrava (Czech Republic)
Title: Statistical Verification on the Performance of Classical Portfolio Optimization Models
Abstract: In the financial investment activities, the strategies made by the portfolio managers are expected to earn higher returns than the random investments by investors who lack of professional knowledge. To make the statistical verification on the historical performance of classical portfolio optimization models, we compare the performance of the strategy portfolios with that of the random-weights portfolios. From the empirical results, we find the chosen risk-minimizing portfolios obtained from the in-sample period does not guarantee its lowest out-of-sample risk for the same risk measure, however, under the considerations of the estimation errors of future returns, we also find that the strategies obtained from the mean-variance model incorporated with the fuzzy estimation method outperform the strategies obtained from the classical mean-variance model.