# Introductory Nonparametrics

:
( 0 )
87 pages
Langue:
English
Introductory Nonparametrics gently introduces the reader to nonparametrics by describing some simple tests, some tests of the most frequently encountered experimental designs, and permutation testing.
Dernière publication
A propos de l'auteur

John Rayner is currently Honorary Professorial Fellow at the Centre for Statistical and Survey Methodology, School of Mathematics and Applied Statistics, University of Wollongong, NSW, Australia and Conjoint Professor of Statistics at the University of Newcastle in NSW, Australia. He served as Profess...

Description
Content
Reviews

Introductory Nonparametrics gives a gentle introduction to nonparametric hypothesis testing. It describes some simple tests, such as the sign and runs tests, and the Kruskal-Wallis, Friedman and Durbin tests, tests of the most frequently encountered experimental designs, the completely randomised, randomised block and balanced incomplete block design respectively. Permutation testing, a fundamental nonparametric tool in its own rite, is introduced to calculate p-values for the tests discussed. A companion text gives detail of R code used throughout Introductory Nonparametrics.

1. A First Perspective on Nonparametric Testing
1. What are nonparametric methods?
2. The sign tests
3. Runs tests
4. The median test
5. The Wilcoxon tests
2. Nonparametric Testing in the Completely Randomised, Randomised Blocks and Balanced Incomplete Block Designs
1. Introduction and outline
2. The Kruskal-Wallis test
3. The Friedman test
4. The Durbin test
5. Relationships of Kruskal-Wallis, Friedman and Durbin tests with ANOVA F tests
6. Orthogonal contrasts: Page and umbrella tests
3. Permutation Testing
1. What is permutation testing and why it is important?
2. Nonparametric multifactor ANOVA when the levels of the factors are unordered
3. Revisiting some previous examples
This is an excellent introductory nonparametrics text. Explanations are clear, real data is often used and there are detailed solutions to the chapter exercises. Of particular note are the use of the free R software, many examples of the use of orthogonal polynomials, discussion of permutation tests and new nonparametric ANOVA. These 4 features are either not covered or covered only briefly in alternative texts. I note the author intends a companion advanced nonparametric text and perhaps in that advice on how to handle missing values and examples of data plots could be given. D.J. Best, PhD
18 septembre 2016 à 07:34
This book is a must for a great introduction to non-parametric testing by one of the masters of the field. The book describes each of the key testing procedures with easy to follow examples.
8 septembre 2016 à 22:23