Synopses & Reviews
Guided by problems that frequently arise in actual practice, James Higgins' book presents a wide array of nonparametric methods of data analysis that researchers will find useful. It discusses a variety of nonparametric methods and, wherever possible, stresses the connection between methods. For instance, rank tests are introduced as special cases of permutation tests applied to ranks. The author provides coverage of topics not often found in nonparametric textbooks, including procedures for multivariate data, multiple regression, multi-factor analysis of variance, survival data, and curve smoothing. This truly modern approach teaches non-majors how to analyze and interpret data with nonparametric procedures using today's computing technology.
Synopsis
Guided by problems that frequently arise in actual practice, James Higgins
About the Author
James J. Higgins is Professor of Statistics at Kansas State University and Fellow of the American Statistical Association. He is the co-author of the Duxbury textbook CONCEPTS IN PROBABILITY AND STOCHASTIC MODELING with Sallie Keller-McNulty and he is author of INTRODUCTION TO MODERN NONPARAMETRIC STATISTICS as well as having over 80 scientific publications to his credit. In addition, he is a statistical consultant for Kansas State Research and Extension. His research interests include nonparametric statistics and reliability theory.
Table of Contents
1. ONE-SAMPLE METHODS. Preliminaries. A Nonparametric Test and Confidence Interval for the Median. Estimating the Population CDF and Quantiles. A Comparison of Statistical Tests. 2. TWO-SAMPLE METHODS. A Two-Sample Permutation Test. Permutation Tests Based on the Median and Trimmed Means. Random Sampling the Permutations. Wilcoxon Rank-Sum Test. Wilcoxon Rank-Sum Test Adjusted for Ties. Mann-Whitney Test and a Confidence Interval. Scoring Systems. Test for Equality of Scale Parameters and an Omnibus Test. Selecting Among Two-Sample Tests. Large Sample Approximations. Exercises. 3. K-SAMPLE METHODS. K-Sample Permutation Tests. The Kruskal-Wallis Test. Multiple Comparisons. Ordered Alternatives. Exercises. 4. PAIRED COMPARISONS AND BLOCKED DESIGNS. Paired Comparison Permutation Test. Signed-Rank Test. Other Paired-Comparison Tests. A Permutation Test for a Randomized Complete Block Design. Friedman's Test for a Randomized Complete Block Design. Ordered Alternatives for a Randomized Complete Block Design. Exercises. 5. TESTS FOR TRENDS AND ASSOCIATION. A Permutation Test for Correlation and Slope. Spearman Rank Correlation. Kendall's Tau. Permutation Tests for Contingency Tables. Fisher's Exact Test for a 2 „e 2 Contingency Table. Contingency Tables With Ordered Categories. Mantel-Haenszel Test. Exercises. 6. MULTIVARIATE TESTS. Two-Sample Multivariate Permutation Tests. Two-Sample Multivariate Rank Tests. Multivariate Paired Comparisons. Multivariate Rank Tests for Paired Comparisons. Multi-response Categorical Data. Exercises. 7. ANALYSIS OF CENSORED DATA. Estimating the Survival Function. Permutation Tests for Two-Sample Censored Data. Gehan's Generalization of the Mann-Whitney-Wilcoxon Test. Scoring Systems for Censored Data. Tests Using Scoring Systems for Censored Data. Exercises. 8. NONPARAMETRIC BOOTSTRAP METHODS. The Basic Bootstrap Method. Bootstrap Intervals for Location-Scale Models. BCA and Other Bootstrap Intervals. Correlation and Regression. Two-Sample Inference. Bootstrap Sampling from Several Populations. Bootstrap Sampling for Multiple Regression. Multivariate Bootstrap Sampling. Exercises. 9. MULTIFACTOR EXPERIMENTS. Analysis of Variance Models. Aligned Rank Transform. Testing for Lattice-Ordered Alternatives. Exercises. 10. SMOOTHING METHODS AND ROBUST MODEL FITTING. Estimating the Probability Density Function. Nonparametric Curve Smoothing. Robust and Rank-Based Regression. Exercises. TABLES. REFERENCES.