Constrained portfolio optimization via Artificial Gorilla Troops: Benchmarking against swarm-intelligence metaheuristic algorithms

Abstract

Over the years, the challenge of portfolio optimization has gained increasing attention from scientists and experienced investors, where maximizing returns with the least possible risk is the major goal. In recent times, there has been a significant surge in interest in metaheuristic algorithms across various industries. In this study, we propose the use of the Artificial Gorilla Troops Optimizer (AGTO) for portfolio optimization. We evaluate its effectiveness and compare it against sixteen other swarm intelligence-based metaheuristic algorithms under varying population and epoch sizes. Our evaluation is based on the Sharpe ratio, utilizing a portfolio composed of stocks from the Dow Jones Industrial Average. The results indicate that AGTO demonstrates strong potential as an effective method for optimal portfolio selection. In addition, this work provides valuable insights into metaheuristic algorithms that have seen relatively limited application in the existing portfolio optimization literature.

Keywords: Artificial Gorilla Troop Optimizer, Constrained optimization, Metaheuristic algorithms, Portfolio optimization, Sharpe ratio.