Dynamic Portfolio Selection with Higher Moments Risk Based on Polynomial Goal Programming

In the presence of higher moments risk, the portfolio selection entails considering competing and conflicting objectives, such as both its expected returns and skewness, and minimizing its variance and kurtosis. At the same time, due to time-varying of higher moments risk, it is necessary to consider dynamic higher moments risk measurement. This article discusses multivariate GARCHSK model based on independent component analysis in the first place. Then we propose a dynamic portfolio selection model with higher moments risk by incorporating the multiple conflicting objectives into a polynomial goal programming problem, where investor's preferences can be designed freely. In the end, empirical analysis is conducted on international stock markets.