This book concerns adaptive tests of significance, which are statistical tests that use the data to modify the test procedures. The modification is used to reduce the influence of outliers. These adaptive tests are attractive because they are often more powerful than traditional tests, and they are also quite practical since they can be performed quickly on a computer using R code or a SAS macro. This comprehensive book on adaptive tests can be used by students and researchers alike who are not familiar with adaptive methods. Chapter 1 provides a gentle introduction to the topic, and Chapter 2 presents a description of the basic tools that are used throughout the book. In Chapters 3, 4, and 5, the basic adaptive testing methods are developed, and Chapters 6 and 7 contain many applications of these tests. Chapters 8 and 9 concern adaptive multivariate tests with multivariate regression models, while the rest of the book concerns adaptive rank tests, adaptive confidence intervals, and adaptive correlations. The adaptive tests described in this book have the following properties: the level of significance is maintained at or near alpha; they are more powerful than the traditional test, sometimes much more powerful, if the error distribution is long-tailed or skewed; and there is little power loss compared to the traditional tests if the error distribution is normal. Additional topical coverage includes: smoothing and normalizing methods; two-sample adaptive tests; permutation tests with linear models; adaptive tests in linear models; application of adaptive tests; analysis of paired data; adaptive multivariate tests; analysis of repeated measures data; rank-based approaches to testing; adaptive confidence intervals; and adaptive correlation.
还没人写过短评呢