Testing uniformity-characterization based approach

Abstract

The uniform distribution is one of the most used distribution in statistical modeling and computer science. Therefore ensuring that the data come from a uniform distribution is of huge importance. Here we present a review of classical and several recent characterization based uniformity tests and their adaptations for testing the composite null hypothesis of rectangular distributions on arbitrary support. Also, some applications in time series will be presented. The presented tests will be compared with respect to several criteria.