Making Decision on Selection of Optimal Stock Portfolio Employing Meta Heuristic Algorithms for Multi-Objective Functions Subject to Real-Life Constraints

The purpose of this study is to utilize data envelopment analysis and metaheuristic algorithms to make investment decisions and select an optimal stock portfolio, considering real-life constraints and multi-objective functions. The statistical population for this research comprises 183 selected companies from the Tehran Stock Exchange involved in capital decision-making and optimal capital composition. Eventually, 42 companies were identified as justifiable investment options. After assessing the risk and return of efficient companies, the study formulated a multi-objective model based on the investor s budget limitations, requirements, and expectations to determine the investment composition. To achieve optimal decisions, a modified genetic metaheuristic algorithm and MATLAB software with dual operators were employed. Sensitivity analysis revealed that eliminating the risk minimization function enhanced the decision s return level but increased risk. Conversely, eliminating the maximizing return function improved decision-making risk but reduced investment return. Eliminating investment requirements and expectations increased returns and investment risk while involving more companies in the optimal investment portfolio.