E-optimization Capable Neural Network for Portfolio Selection Chance-Constrained Programming Model

This paper presents a neural network model to solve the portfolio selection problem when security returns are uncertain variables. Two types of portfolio selectionprogramming models based on uncertain measure are provided according to uncertain theory. The main idea is to replace theportfolio selection models to their crisp equivalents when thereturn rates are adopted some special uncertain variables such as linear uncertain variable, trapezoidal uncertain variable andnormal uncertain variable. According to the saddle point theorem, optimization theory, convex analysis theory, Lyapunov stability theory and LaSalleinvariance principle, the equilibrium point of the proposed neural network is proved to be equivalent to the optimal solutionof the original problem. It is also shown that the proposed neural network model is stable in the sense of Lyapunov and it isglobally convergent to an exact optimal solution of the portfolioselection problem with uncertain returns. An illustrative example is provided to show the feasibility and the efficiency of the proposed method in this paper. The simulation is conducted on Matlab software, and this process can be simulated in internet.

کلیدواژه ها:

Portfolio selection ، Uncertain variable ، Neural network ، Matlab ، e-Optimization