كاربرد الگوريتم ژنتيك در انتخاب يك مجموعه دارايي از سهام بورس اوراق بهادار

Application of Genetic Algorithm in Portfolio Selection Problem

One of the classical applications of operation research in investment decision making is the portfolio selection problem. In this problem a fixed sum of money is to be spread among different investments and there is a risk associated with the rate of return on each investment. The object of the portfolio selection problem is to determine how much money should be allocated to each investment to maximize the total expected return and minimize the portfolioʹs risk. Since there is no specific algorithm to find an optimal feasible solution for large scale portfolio problems, in this paper two genetic algorithms are developed to find a near optimal solution. In the first algorithm the selection of investments is determined and in the second one the weight of each investment in the portfolio is calculated. Finally, the two algorithms have been applied successfully to the portfolio of stocks of the Tehran Stock Exchange with more than 200 stocks.