Quantum-inspired meta-heuristic approaches for a constrained portfolio optimization problem

Abstract

Portfolio optimization has long been a challenging proposition and a widely studied topic in finance and management. It involves selecting and allocating the right assets according to the desired objectives. It has been found that this nonlinear constraint problem cannot be effectively solved using a traditional approach. This paper covers and compares quantum-inspired versions of four popular evolutionary techniques with three benchmark datasets. Genetic algorithm, differential evolution, particle swarm optimization, ant colony optimization, and their quantum-inspired incarnations are implemented, and the results are compared. Experiments have been carried out with more than 10 years of stock price data from NASDAQ, BSE, and Dow Jones. This work proposes several enhancements to allocate funds efficiently, such as improved crossover techniques and dynamic and adaptive selection of parameters. Furthermore, it is observed that the quantum-inspired techniques outperform the classical counterparts.