Asymptotic efficiency of nonparametric tests

Choosing the most efficient statistical test is one of the basic problems of statistics. Asymptotic efficiency is an indispensable technique for comparing and ordering statistical tests in large samples. It is especially useful in nonparametric statistics where there exist numerous heuristic tests such as the Kolmogorov-Smirnov, Cramer-von Mises, and linear rank tests. This monograph discusses the analysis and calculation of the asymptotic efficiencies of nonparametric tests. Powerful methods based on Sanov's theorem together with the techniques of limit theorems, variational calculus, and nonlinear analysis are developed to evaluate explicitly the large deviation probabilities of test statistics. This makes it possible to find the Bahadur, Hodges-Lehmann, and Chernoff efficiencies for the majority of nonparametric tests for goodness-of-fit, homogeneity, symmetry, and independence hypotheses. Of particular interest is the description of domains of the Bahadur local optimality and related characterization problems, based on recent research by the author. The general theory is applied to a classical problem of statistical radio physics: signal detection in noise of unknown level. Other results previously published only in Russian journals are also published here for the first time in English.