Abstract

The paper addresses the constrained mean-semivariance portfolio optimization problem with the support of a novel multi-objective evolutionary algorithm (n-MOEA). The use of semivariance as the risk quantification measure and the real world constraints imposed to the model make the problem difficult to be solved with exact methods. Thanks to the exploratory mechanism, n-MOEA concentrates the search effort where is needed more and provides a well formed efficient frontier with the solutions spread across the whole frontier. We also provide evidence for the robustness of the produced non-dominated solutions by carrying out, out-of-sample testing during both bull and bear market conditions on FTSE-100.

Highlights

  • Portfolio optimization is the process of choosing the assets and their proportions, so that it is attained the maximum profitability for the risk undertaken

  • The use of semivariance as the risk quantification measure and the real world constraints imposed to the model make the problem difficult to be solved with exact methods

  • Thanks to the exploratory mechanism, novel multi-objective evolutionary algorithm (n-MOEA) concentrates the search effort where is needed more and provides a well formed efficient frontier with the solutions spread across the whole frontier

Read more

Summary

Introduction

Portfolio optimization is the process of choosing the assets and their proportions, so that it is attained the maximum profitability for the risk undertaken. MOEAs techniques applied to the portfolio selection problem have become increasingly popular relatively recently. Because they provide a fast and reliable way of calculating computationally demanding financial models and why revolutionized the financial modeling research field itself by developing innovative algorithmic approaches for solving difficult financial problems that in many cases cannot be solved with exact methods. Incorporate a number of constraints into the optimization model, but do not examine how these constraints affect the evolutionary search process and the efficient frontier formulation. Our paper proposes a bi-objective return semivariance portfolio optimization model that incorporates a number of real world constraints such as cardinality constraints, floor and ceiling constraints, non-negativity constraint and budget constraint and analyzes their effects on the efficient frontier formulation.

Problem Definition
Constraints to the Problem
Pareto Optimality Definitions
The Proposed Methodology
Generation of New Population
The Mutation Mechanism
The Crossover Mechanism
Efficient Frontier Formulation
Robustness of n-MOEA
Findings
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call