Optimizing fuzzy portfolio selection problems by parametric quadratic programming

  • Authors:

  • Xiao-Li Wu;Yan-Kui Liu

  • Affiliations:

  • Key Laboratory in Machine Learning and Computational Intelligence, College of Mathematics & Computer Science, Hebei University, Baoding, China 071002;Key Laboratory in Machine Learning and Computational Intelligence, College of Mathematics & Computer Science, Hebei University, Baoding, China 071002

  • Venue:
  • Fuzzy Optimization and Decision Making
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper develops a robust method to describe fuzzy returns by employing parametric possibility distributions. The parametric possibility distributions are obtained by equivalent value (EV) reduction methods. For common type-2 triangular and trapezoidal fuzzy variables, their reduced fuzzy variables are studied in the current development. The parametric possibility distributions of reduced fuzzy variables are first derived, then the second moment formulas for the reduced fuzzy variables are established. Taking the second moment as a new risk measure, the reward-risk and risk-reward models are developed to optimize fuzzy portfolio selection problems. The mathematical properties of the proposed optimization models are analyzed, including the analytical representations for the second moments of linear combinations of reduced fuzzy variables as well as the convexity of second moments with respect to decision vectors. On the basis of the analytical representations for the second moments, the reward-risk and risk-reward models can be turned into their equivalent parametric quadratic convex programming problems, which can be solved by conventional solution methods or general-purpose software. Finally, some numerical experiments are performed to demonstrate the new modeling ideas and the efficiency of solution method.