The classical theories of testing and estimation are based on the assumption that the underlying distributions are normal or belong to some other parametric family. This chapter considers the development of new methodologies appropriate to models in which the distributions are not so restricted. Following Wolfowitz (1942), such models are called nonparametric.
KeywordsWilcoxon Test Influence Function Breakdown Point Nonparametric Inference Permutation Version
Unable to display preview. Download preview PDF.