انتخاب و بهينه سازي سبد سهام با استفاده از الگوريتم ژنتيك بر اساس تعاريف متفاوتي از ريسك

Selecting and Optimizing the Portfolio Using the Genetic Algorithm Based on Different Definitions of Risk Portfolio

In finance, a portfolio is an appropriate mix or collection of investments held by an institution or an individual. Selecting a portfolio in order to maximize income is one of the main problems of the investors in the financial markets. The existing methods are not enough efficient on the optimal selection of portfolio, so in order to solve this problem some initiative algorithms have been recently considered. Genetic algorithm is one of the initiative algorithms which can successfully optimize the portfolio with regarding to the different levels of risk. The purpose of this research is selecting and optimizing the portfolios based on the different levels of risk. In order to achieve this goal, two genetic algorithms were designed. In the process of designing the needed algorithms, two basic models (i.e. mean-variance model of Markowitz and mean-semi variance model) were considered. In order to be more efficient, some real world constraints were added to the designed algorithms. MATLAB7 for designing genetic algorithms and single sample t-test for testing the research hypotheses were used. Statistical community of this research was 146 companies operating in Tehranʹs Stock Exchange and its time frame was from 1380 to 1387.The research results revealed that the designed genetic algorithm has enough optimality and stability on the case of continues repeats. According to the results, there is no meaningful difference between using two models (i.e. mean-variance model of Markowitz and mean-semi variance model). By using the designed genetic algorithms, we believe the investors will be able to select an optimal portfolio.