PSO applied to different risk measures in portfolio selection problems

Abstract

In this thesis are presented some of the principal risk measures that are utilized in the financial field and in portfolio construction. The aim of this work is to establish if one of the selected risk measures can be considered better than the others into the construction of financial portfolios. Starting from the well-known Harry Markowitz’s modern portfolio theory (MPT), these risk measure will substitute the Variance in the computation of the optimal portfolios. A metaheuristic based algorithm (Particle Swarm Optimization) will be used in order to solve the optimization problem as proposed in the MPT. In order to obtain a fair comparison between the resulting portfolios, they will be examined using different indexes that will be presented as well