Particle Swarm Algorithm: An Application on Portfolio Optimization

Abstract

Optimization is discovering an alternative with the most cost-effective or highestachievable performance under the given constraints, by maximizing desired factors and minimizing undesired ones. Portfolio optimization in finance depends on selecting assets from an opportunity set which yields highest expected return on each level of portfolio risk. Optimization algorithms based on natural events are called heuristic algorithms. The particle swarm optimization (PSO) is a population-based heuristic optimization technique. The technique is inspired by the ability of animals such as birds and fish to adapt to their environment by applying a “sharing of knowledge” approach, to find rich food sources and to avoid hunting. This chapter focuses on portfolio selection problems and shows how to manage financial portfolios using a particle swarm optimization (PSO) technique which is a heuristic algorithm. In order to better understand the subject, the technique has been evaluated in Istanbul Stock Exchange for three transportation sector stocks.