BI Statistical Methods

In this ground-breaking work, Peter Walley constructs and justifies a radically rational theory of statistical inference. He derives, from fundamental principles, new statistical methods that are coherent and objective and correct crucial failings of standard methods. Volume 1 presents the foundations and the general theory of statistical inference. Volume 2 applies the theory to derive practical methods for analysing common types of statistical data. The book unifies the wide-ranging, highly original research produced by the author during 1980-2005, most of which has been unpublished until now. The methodology developed in the book provides a general, systematic and coherent way to analyse statistical problems. The methods apply to all kinds of statistical problem and all types of data; inference about parameters, predictive inference, and nonparametric inference; informative or uninformative prior models; point estimation, interval estimation, hypothesis testing, density estimation, and statistical decision problems. They also enable a unified treatment of robustness. The book should be of great interest to everyone who uses or studies statistical inference. The results have profound implications for a wide range of disciplines in which statistical methods are important, including econometrics, decision theory, inductive logic, philosophy of science, psychology, and clinical trials. Libraries and bookstores can order the book through the Ingram database (ipage). For more information about the book, including an extended outline, table of contents, and the full text of chapter 1, visit www.pspress.net/bimethods/