Metaheuristics for portfolio optimization

Abstract

Portfolio optimization refers to allocating an amount of investors’ wealth to different assets in order to satisfy the investors’ preferences for return and risk. We address the portfolio optimization problem with real-world constraints, where traditional optimization methods fail to efficiently find an optimal or near-optional solution. Hence, we design a modified cuckoo search (MCS) metaheuristic for finding good sub-optimal portfolios. Cuckoo search was inspired by the brood parasitism of cuckoo species by laying their eggs in the nests of other host birds. Our implementation explores the search space using Levy flights and allocates the good sub-optimal distribution of investment weights for a chosen set of assets. The MCS results show a clear improvement in comparison with previously published results, based on Markowitz and Sharpe models.