Portfolio selection using genetic algorithms

Abstract

The study of portfolio optimization pioneered by Harry Markowitz in his mean-variance (E-V) model is undoubtedly a popular research area in finance. The combination of different assets in a master asset often called a portfolio is more profitable than holding a single asset, since the risk of investment is diversified away without necessarily degrading the expected return, given that the right choice of constituent assets is made and proper weights are allocated. In this paper, one of the nature-inspired evolutionary algorithms – a Genetic Algorithms (GA) was used in solving the portfolio selection problem (PSP). Based on a real dataset from a popular stock market, the performance of the algorithm in relation to those obtained from one of the popular quadratic programming (QP) solver – MINOS 5.5 was reported.

Keywords: Portfolio selection, genetic algorithms, efficient frontier, stock market

International Journal of Natural and Applied Sciences, 5(4): 365-373, 2009