Stochastic Dominance, Dynamic Programming and Bayes Filtering: Applications in NPV and NPVaR

Applications of three computational methods in the evaluation of NPV and NPV at risk (NPVaR) of a project are studied. These methods are stochastic dominance, dynamic programming and Bayes filtering. First, the definition of NPVaR is related to the stochastic dominance as a maximization problem, then the dynamic programming method is used to solve the maximization in the presence of an investor belief parameter. To enter this parameter to the problem, the Bayes filter method is applied. Literatures is reviewed about these three methods. The sensitivity analysis of the results to the different utility functions as well as to the continuous compounding methods are also studied. A real data set is presented. Finally, a conclusion section is also given.