Selection of Appropriate Portfolio Optimization Strategy

Abstract

Portfolio optimization is a critical task in financial management, aiming to maximize expected returns while minimizing risk. This study compares the performance of Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Dynamic Programming (DP), and Differential Evolutionary Algorithm (DEA) in optimizing portfolios in the NIFTY 50 market. Using daily stock data from March 2023 to May 2023, we evaluate the algorithms based on performance metrics including the Sharpe ratio, expected return, volatility and Sortino ratio. Then an integrated multi-criteria decision making (MCDM) framework of Logarithmic Percentage Change-driven Objective Weighting (LOPCOW) and Compromise Ranking of Alternatives from Distance to Ideal Solution (CRADIS) methods has been used to compare the evolutionary algorithms for portfolio optimization. The results show that PSO outperforms the other algorithms in terms of Sharpe Ratio and Expected Return, while DEA exhibits the lowest portfolio risk. Furthermore, the efficient frontier analysis confirms PSO's ability to generate portfolios with higher expected returns at the same risk level. This research highlights the effectiveness of optimization algorithms in portfolio management and provides valuable insights for investors and portfolio managers in maximizing returns and managing risks.