Quantum-inspired Jaguar Algorithm for Real-World Financial Problems

Quantum-inspired algorithms simulate quantum characteristics on classical computers to accelerate the solution of complex real-world optimization problems. In the era before general-purpose quantum computers become practical, they offer a promising and accessible alternative for tackling such challenges. Among optimization problems, combinatorial types are particularly challenging due to discrete spaces, large scales, and strong variable dependencies. This study proposes a novel quantum-inspired algorithm named the quantum-inspired jaguar algorithm (QJA), specifically designed to address highly discrete and structurally complex combinatorial optimization problems. QJA integrates the adaptive hunting mechanism of the jaguar algorithm with the core concept of quantum-inspired tabu search algorithm, which guides the search by moving toward the best solution while avoiding the worst one. QJA dynamically adjusts its parameters based on historical information to accelerate convergence and efficiently explore the solution neighborhood. In addition, the algorithm incorporates an entangled local search mechanism to further enhance solution quality. QJA is evaluated on real-world portfolio optimization using data from the Toronto Stock Exchange. Results show that it outperforms other quantum-inspired algorithms in both stability and solution quality, demonstrating superior convergence and practical applicability.