Multi-period Robust Portfolio Selection Model with Bat Algorithm

Xing Yu
2017 International Journal of Control and Automation  
This paper proposes a portfolio selection model in which the methodologies of robust optimization are used for the maximization of the terminate wealth under the constraints of each stage risk described as absolute deviation of not more than the given levels. In place of stochastic programming, it is used of techniques of robust optimization to deal with uncertainty. Moreover, we succeeded in solving the multi-period portfolio selection model with use of the bat algorithm (BA). Numerical
more » ... show that there are two important features of our work. One is that the yield curve is smooth, less fluctuation. So, it is of less psychological impact on investors, cash flow smooth. Another, BA applied to solve the model is feasible and it is more effective than genetic algorithm. MV model is that the parameters are assumed to be certain according to the history data. But it is a fact that we should make decision for the future, but future is not be replaced by history, and the future is uncertain. So the optimization process leads to solutions heavily on the parameters perturbations. In this context, we take account in uncertainty in optimization model and make worst estimations about the unknown information which is so-called robust optimization technique. Anna deals with a portfolio selection model in which the methodologies of robust optimization are used for the minimization of the conditional value at risk of a portfolio of shares. Shapiro considers the adjustable robust approach to multistage optimization, for which we derive dynamic programming equations. The rest of the paper is organized as follows. Section 2 describes the multi-period mean deviation optimization problem. In section 3, we introduce robust framework of the model. In section 4 the BA is presented. Section 5 presents the numerical example and results to show the practically and efficiency of our model and algorithm. Finally, section 6 conclusions the paper.
doi:10.14257/ijca.2017.10.5.03 fatcat:utabpxj7ivbizafqds3huqlpdu