Limit theorems are at the heart of the mathematical statistics and, in particular, of the theories of estimation and of statistical tests. Among the limit theorems, strong approximations play an important role. Here, we focus on the Komlos-Major-Tusnhdy (KMT) approximation, also known as Hungarian representation. The KMT approximation provides a remarkable, mathematically tractable representation of the empirical process by a Brownian bridge constructed on the same probability space. We describe the KMT approximation, particularly as it relates to other forms of approximation, and to review some of its applications, especially in the nonparametric inference

Keywords