Permutation Methods
A Distance Function Approach
Buch in deiner Nähe kaufen
Beschreibung
Most commonly-used parametric and permutation statistical tests, such as the matched-pairs t test and analysis of variance, are based on non-metric squared distance functions that have very poor robustness characteristics. This second edition places increased emphasis on the use of alternative permutation statistical tests based on metric Euclidean distance functions that have excellent robustness characteristics. These alternative permutation techniques provide many powerful multivariate tests including multivariate multiple regression analyses. In addition to permutation techniques described in the first edition, this second edition also contains various new permutation statistical methods and studies that include resampling multiple contingency table analyses, analysis concerns involving log-linear models with small samples, an exact discrete analog of Fisher’s continuous method for combining P-values that arise from small data sets, multiple dichotomous response analyses, problems regarding Fisher’s Z transformation for correlation analyses, and multivariate similarity comparisons between corresponding multiple categories of two samples. Paul W. Mielke, Jr. is Professor of Statistics at Colorado State University, and a fellow of the American Statistical Association. Kenneth J. Berry is Professor of Sociology at Colorado State University.
Classical statistical analysis depends on many assumptions about the nature of data and the population from which the samples were drawn. Permutation tests depend only on the data gathered and, although computationally intensive, provide powerful alternatives to conventional analysis techniques. Most commonly-used parametric and permutation statistical tests, such as the matched-pairs t test and analysis of variance, are based on non-metric squared distance functions that have very poor robustness characteristics. This Second Edition places increased emphasis on the use of alternative permutation statistical tests based on metric Euclidean distance functions that have excellent robustness characteristics. These alternative permutation techniques provide many powerful multivariate tests including multivariate multiple regression analyses.
Makes a variety of powerful data analytic tools easily available to practitioners New material has been added to the second edition
Autor*in
Paul W. Mielke
Themen in »Permutation Methods«
Excel Permutation statistics Regression analysis Resampling Variance analysis of variance best fit correlation data-dependent methods goodness-of-fit multiple contingency tables multivariate analysis
Stimmen zu »Permutation Methods«
From the Reviews:
"[T]his is a nicely written book that contains many important and useful topics, and I am certan that many pracitioners and researchers will find the new edition beneficial." (Technometrics, May 2008, Vol. 50 No. 2)
“This is a very well-written text that extensively covers permutation-based tests in a general framework. It has been revised and extended by nearly 100 pages since the 2001 edition. …This book is packed with real-data examples and dozens of simulation studies exploring the properties of permutation-based tests and contrasting them with their typical parametric ‘competitors.’ The authors do not shy away from presenting the mathematical underpinnings of the methods, and do so in a very transparent and easy-to-follow manner so there is sufficient detail to implement the methods in your favorite software … . That is not a concern if you are familiar with FORTRAN-77, as the authors have provided over 100 FORTRAN programs an associated datasets for download in Unix-compatible and Windows-compatible format. These well-commented programs are briefly described in Appendix A with subsections organized by chapter. …Permutation Methods is a superb book that is highly recommended.” ( Journal of the American Statistical Association, Dec. 2009, Vol. 104, No. 488)()