Hybrid meta-heuristic algorithm for multi-objective portfolio optimization by fuzzy programming

Subject Areas : Financial engineering

Javad Behnamian ¹ , Mohammad Moshrefi ²

1 - Assistant Professor, Department of Industrial Engineering, Faculty of Engineering, Bu-Ali Sina University, Hamedan, Iran
2 - Msc, Department of Industrial Engineering, Faculty of Engineering, Bu-Ali Sina University, Hamedan, Iran

Received: 2016-07-31 Accepted : 2017-10-18 Published : 2017-04-21

Keywords: Fuzzy Mathematical Programming, Stock Portfolio, Bvnysvn Method, variable neighborhood search,

Abstract :

Portfolio selection is one of the problems which is very important for the investors in the exchange market. In order to invest on several kinds of shares instead of a specific one so that they can earn maximum profit through a certain level of risk or have minimum risk with a certain level of profit. What have been introduced concerning financial calculations and the choice of portfolio or selection of an investment up to now, in some ways, determines priority in accordance with degree of risk and rate of return respectively; so that the investor can select his favourite portfolio through financial facilities and other confronting policies. When the investor is faced with different properties, he should make decisions about the number of properties as well as the amount of investment. So in some ways, he faces uncertain decisions in his choices. We are considering fuzzy concepts in the discussion of portfolio selection optimization in order to pursue this uncertainty. Then by using Bonison method, we determine priority and preference among the portfolios so that the investor can decide without confusion and finally through introducing combined metaheuristic algorithm for variable neighbourhood search and genetics, can optimize the resulting model of previous process and comparing it with other solving algorithms.