Harris hawks optimisation algorithm for the multi-objective portfolio optimisation with transaction costs

Résumé

Abstract The portfolio optimisation problem (POP) is a financial engineering problem where an investor is faced with the issue of allocating wealth amongst a set of assets so as to obtain a fair balance between the expected return and risk of a portfolio. In this paper we consider the multi-objective POP subject to the cardinality, quantity and pre-assignment constraints and study the cases when transaction costs are omitted and considered in the model. In order to solve this problem, the Harris Hawks Optimisation (HHO) algorithm is employed. Given the combinatorial nature of the constrained POP, a uniform crossover operator is incorporated in the exploration phase of the HHO that deals with the combination of assets for each candidate portfolio. We first test the applicability of the algorithm to the unconstrained problem and show that it is capable of obtaining highly accurate results. Then the suitability of the modified HHO is tested with seven publicly available datasets and compared against well-known heuristics. Five performance metrics were used to evaluate the performance of the algorithms and the results obtained proved the efficacy and superiority of the HHO over the other methods even in the presence of transaction costs.

Citer ce document

Accès au document

Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter

Auteur(s)

Statistiques

Consultations : 1

Téléchargements : 0