A comparison on particle swarm optimization and genetic algorithm performances in deriving the efficient frontier of stocks portfolios based on a mean-lower partial moment model

In this paper, a portfolio optimization model on the basis of the risk measure of lower partial moment of the first order is discussed. Two meta-heuristic methods of particle swarm optimization and genetic algorithm performances are applied and compared from different aspects to derive the stocks portfolios efficient frontier. The data belongs to the monthly returns of 20 randomly selected and approved stocks in the New York Stock Exchange for the financial period of 2005–2011. The results prove that both algorithms are quite efficient in solving the mean-lower partial moment of the first order model with the particle swarm optimization being superior.