A mean-variance-skewness model: Algorithm and applications

We will show that a mean-variance-skewness portfolio optimization model, a direct extension of the classical mean-variance model can be solved exactly and fast by using the state-of-the-art integer programming approach. This implies that we can now calculate a portfolio with maximal expected utility for any decreasing risk averse utility function. Also, we will show that this model can be used as a practical tool for constructing a portfolio when the asset returns follow skewed distribution. As an example, we apply this model to construct an index plus alpha portfolio.