Study of portfolio optimization problem based on gravitational search algorithm
Liu Xiao-yon · Jisuanji yingyong yanjiu · 2014
Gravitational search algorithm(GSA) is a novel optimization algorithm based on the law of gravity and mass interactions. In the algorithm,the searcher agents were a collection of masses which interacted with each other based on the Newtonian gravity and the laws of motion. Some study show that GSA can obtain more superior results than PSO in most cases. But the basic gravitational search algorithm was eary to be trapped into local optimum. This paper proposed an improvement weighted algorithm,named gGSA,and strengthed local search capacity of GSA. This algorithm was used to portfolio optimization model based on VaR and solved the portfolio optimization problem. The result demonstrates that the new algorithm is feasible and effective.