A novel Meta-Heuristic method for solving an extended Markowitz Mean- Variance portfolio selection model

This paper presents a novel Meta-Heuristic method for solving an extended Markowitz Mean–Variance portfolio selection model. The extended model considers Value-at-Risk (VaR) as risk measure instead of Variance. Depending on the method of VaR calculation its minimizing methodology differs. if we use Historical Simulation which is applied in this paper then the curve would be non-convex.
On the other hand the Mean-VaR model here includes three sets of constraints: bounds on holdings, cardinality and minimum return which cause a Mixed Integer Quadratic Programming Problem. The first set of constraints guarantee that the amount invested (if any) in each asset is between its predetermined upper and lower bounds. The cardinality constraint ensures that the total number of assets selected in the portfolio’s equal to a predefined number.
Because of above mentioned reasons, in this paper, we propose a new Meta-Heuristic approach based on combined Ant Colony Optimization (ACO) method and Genetic Algorithm (GA). The computational results show that the proposed Hybrid Algorithm has the ability to optimized Mean-VaR portfolio for small portfolio .