An efficient hybrid metaheuristic algorithm for cardinality constrained portfolio optimization

Portfolio optimization with cardinality constraints turns out to be a mixed-integer quadratic programming problem which is proven to be NP-Complete that limits the efficiency of exact solution approaches, often because of the long-running times. Therefore, particular attention has been given to approximate approaches such as metaheuristics which do not guarantee optimality, yet may expeditiously provide near-optimal solutions. The purpose of this study is to present an efficient hybrid metaheuristic algorithm that combines critical components from continuous ant colony optimization, artificial bee colony optimization and genetic algorithms for solving cardinality constrained portfolio optimization problem. Computational results on seven publicly available benchmark problems confirm the effectiveness of the hybrid integration mechanism. Moreover, comparisons against other methods' results in the literature reveal that the proposed solution approach is competitive with state-of-the-art algorithms.