RSD: An R package to calculate stochastic dominance

Stochastic dominance is a classical method for comparing two random variables using their probability distribution functions. As for all stochastic orders, stochastic dominance does not always establish an order between the random variables, and almost stochastic dominance was developed to address such cases, thus extending the applicability of stochastic dominance to many real-world problems. We developed an R package that consists of a collection of methods for testing the first- and second-order (almost) stochastic dominance for discrete random variables. This article describes the package and illustrates these methods using both synthetic datasets covering a range of possible scenarios that can occur, and a practical example where the comparison of discrete random variables using stochastic dominance can be applied to aid decision-making.