Log-optimal portfolio models with risk control of VaR and CVaR using genetic algorithms

  • Authors:

  • Sen Qin

  • Affiliations:

  • Hangzhou Dianzi University, Hangzhou, Zhejiang, China

  • Venue:
  • Proceedings of the first ACM/SIGEVO Summit on Genetic and Evolutionary Computation
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Value-at-risk (VaR) and conditional value-at-risk (CVaR) have become two very popular measures of market risk during the last decade. Log-optimal portfolio problem with risk control of VaR and CVaR is put forward firstly. Then, we propose the portfolio models with VaR and CVaR and prove the existence and uniqueness of the optimal solutions of these two models. We provide a newly genetic algorithm based on real-code strings of assets' returns to overcome the problem of local optima. Finally, an empirical study is carried out to illustrate the optimal solutions of the log-optimal portfolio models with VaR and CVaR. The numeric results indicate that the optimal portfolio of the log-optimal portfolio model with CVaR gives a balance between the investment risk and the return simultaneously, and is more effective than the corresponding portfolios of the VaR model and the mean-variance model.