Год выпуска: 2011 Автор: Magid Maatallah Издательство: LAP Lambert Academic Publishing Страниц: 512 ISBN: 9783844329605
Описание
This book is based on the premise that nonparametric or rank based statistical methods are a superior choice in many data analytic situations. We cover location models, regression models including designed experiments, and multivariate models. Geometry provides a unifying theme throughout much of the development. We emphasize the similarity in interpretation with least squares methods. Basically, we replace the Euclidean norm with a weighted L-1 norm. on the linear model, sections one through seven and section nine present the basic material for estimation, testing and diagnostic procedures for model criticism. Sections two through four of Chapter 4 give extensive development of methods for the one- and two-way layouts. Then, depending on individual tastes, there are several more exotic topics in each chapter to choose from.
