Effects of higher order moments on the newsvendor problem

In this paper, we propose a stochastic programming model for the well-known single-period newsvendor problem by adopting the conditional Value-at-Risk (CVaR) as the risk metric in the objective function. The demand uncertainty is modeled in terms of discrete scenarios that reflect the empirical distributions implied by market demand data. Our numerical results demonstrate that the higher order moments (skewness and kurtosis) of demand have obvious effects on the newsvendor problem, and the stochastic programming framework provides a flexible and effective decision support tool for the newsvendor problem.