Developing a Fuzzy Multibjective Model for Multiperiod Portfolio Optimazation Considering Average Value at Risk

Subject Areas : Financial engineering

Amir  Shiri Ghehi ¹ (department of Finance, Islamic Azad University, Aliabad Katoul Branch)
Hosein  Didehkhani ² (Department of Financial Engineering, Islamic Azad University, Aliabad Katoul, Iran)
kaveh  Khalili Damghani ³ (Department of Industrial Engineering, South-Tehran Branch, Islamic Azad University, Tehran, Iran.)
parviz  Saeedi ⁴ (Department of financial Management, Aliabad katoul Branch, Islamic Azad University, Aliabad katoul , Iran)

Received: 2017-07-17 Accepted : 2017-12-18 Published : 2018-06-22

Keywords: Portfolio optimization, MOPSO algorithm, Average Value at Risk, multi-period portfolio, credibility theory,

Abstract :

The purpose of the present research is to provide a multi-period portfolio optimization model in a fuzzy credibility environment, aimed for end-of-period wealth maximization and risk minimization. The investor’s risk was measured using the Average Value at Risk (AVaR) as a coherent risk measure. The model is designed in such a way that, in addition to considering transaction costs, the investor will have the opportunity to allocate part of his wealth to a risk-free asset. In designing the model, in addition to the cardinality constraints, constraints such as the minimum “proportion entropy” (as the portfolio of diversification degree) and the expected returns of the portfolio in each period are considered. The results of the model running by MOPSO algorithm indicated that the model objectives in the optimum portfolios were better suited than those when the model was run with random weights. The results also indicated that an increase in the portfolio diversification degree reduced the amount of the final wealth.