Optimal Conditional Mean-Variance Portfolio Averaging
Speaker:Zhang Xinyu, Researcher, Academy of Mathematics and Systems Science, Chinese Academy of Sciences
Host:Zhou Xianbo, Professor, Lingnan College
Time and Date:14:30,Dec,18,2023
Venue:Chen Rongjie Lecture Hall (302), Lingnan Hall
Language: Chinese
Abstract:
In this article, we develop a novel portfolio averaging strategy under the conditional mean-variance framework for achieving a desired risk-return trade-off. A series of shrunken candidate portfolios are constructed, which differ from each other in candidate models, target portfolios, weighting matrices and/or penalty parameters. We adopt a modified Mallows-type criterion to determine the weights across these candidate portfolios. Theoretically, we establish the asymptotic optimality of the proposed strategy in the sense of achieving the lowest possible out-of-sample expected utility loss, and also derive the convergence of weights arising from this criterion. Empirically, we illustrate that the proposed strategy compares favorably with 15 alternative strategies across five datasets.
ZHANG Xinyu Professor
Department of Statistics and Finance
Discipline: Probability and statistics
Email:xinyu143@ustc.edu.cn