A Hybrid Intelligent Algorithm for Portfolio Selection using Fuzzy Mean-Variance-Skewness

The most important problem for investors, at the beginning stages of their works, is the way of assigning their investment to one or more different investment alternatives in such a way that with the least possible risk the maximum return become obtainable. In the economic literature this is known as the problem of portfolio selection. This article tries to introduce an efficient way for supporting decision maker in the selection of appropriate portfolio for investment purposes. The portfolio is based upon the mean-variance-skewness with the return of portfio is considered to be fuzzy to match with the world reality more. This article proposes a hybrid intelligent algorithm for finding an optimial or new optimal solution of the problem. Here, authors use Genetic Algorithm to find the right portfolio with the help of neural network and fuzzy computer simulation knowledge. Due to the fact that trained neural network was used the computation time has reduced tremendously in comparison with the straight use of the fuzzy simulation. Authors have used two example problems to demonstrate the efficiency of the proposed algorithm in comparison with other hybrid algorithms from the literature.