F. Salas Molina, D. Pla Santamaria, A. García Bernabeu, J. Reig Mullor
The mean-variance portfolio selection model by Markowitz minimizes the expected variance of returns subject to a given target of returns. To address some of the limitations of the mean-variance model, hierarchical risk parity methods have recently been proposed based on machine learning. Alternative approaches, such as the inverse variance portfolio and the equally weighted portfolio, present the advantages of simplicity. In this paper, we perform a comparative analysis of the mean-variance model, a hierarchical risk parity method, an inverse variance approach, and an equally weighted portfolio. We use Monte Carlo simulation to evaluate the performance of alternative portfolio selection models under virtual scenarios. We pay special attention to the influence of changes in the main features of the virtual scenarios by varying the correlation between the return of assets, adding random shocks, and considering different market trends.
Keywords: Portfolio selection, risk budgeting, clustering, performance, virtual scenarios.
Scheduled
Operations Research Methods and Aplications
November 7, 2023 6:40 PM
HC1: Canónigos Room 1