Barry C. Arnold - Böcker

Since the publication of the first edition over 30 years ago, the literature related to Pareto distributions has flourished to encompass computer-based inference methods. Pareto Distributions, Second Edition provides broad, up-to-date coverage of the Pareto model and its extensions. This edition expands several chapters to accommodate recent results and reflect the increased use of more computer-intensive inference procedures.New to the Second Edition New material on multivariate inequality Recent ways of handling the problems of inference for Pareto models and their generalizations and extensionsNew discussions of bivariate and multivariate income and survival models This book continues to provide researchers with a useful resource for understanding the statistical aspects of Pareto and Pareto-like distributions. It covers income models and properties of Pareto distributions, measures of inequality for studying income distributions, inference procedures for Pareto distributions, and various multivariate Pareto distributions existing in the literature.

Statisticians, probabilists, and mathematicians will all be interested in a expanded version of this classic work on inequalities. The theory of inequalities has applications in virtually every branch of mathematics.

My interest in majorization was first spurred by Ingram aIkin's proclivity for finding Schur convex functions lurking in the problem section of every issue of the American Mathematical Monthly. Later my interest in income inequality led me again to try and "really" understand Hardy, Littlewood and Polya' s contributions to the majori­ zation literature. I have found the income distribution context to be quite convenient for discussion of inequality orderings. The pre­ sent set of notes is designed for a one quarter course introducing majorization and the Lorenz order. The inequality principles of Dalton, especially the transfer or Robin Hood principle, are given appropriate prominence. Initial versions of these notes were used in graduate statistics classes taught at the Colegio de Postgraduados, Chapingo, Mexico and the University of California, Riverside. I am grateful to students in these classes for their constructive critical commentaries. My wife Carole made noble efforts to harness my free form writ­ ing and punctuation. Occasionally I was unmoved by her requests for clarification. Time will probably prove her right in these instances also. Peggy Franklin did an outstanding job of typing the manu­ script, and patiently endured requests for innumerable modifications.

Bounds on moments of order statistics have been of interest since Sir Francis Galton (1902) flrst addressed the problem of fairly dividing flrst and second prize money in a competition. The present compendium of results represents our effort to sort the plethora of results into some semblance of order. We have tried to assign priority for results appropriately. We will cheerfully accept corrections. Omissions of interesting results have inevitably occurred. On this too we await (cheerful) corrections. We are grateful to Peggy Franklin (University of California), Janet Leach, Domenica Calabria and Patsy Chan (McMaster University) who shared the responsibility of typing the manuscript. The flnal form of the manuscript owes much to their skill and patience. Barry C. Arnold Riverside, California U. S. A. N. Balakrishnan Hamilton, Ontario Canada November, 1988 Table of Contents Chapter 1: TIlE DISTRIBUTION OF ORDER STATISTICS Exercises 4 Chapter 2: RECURRENCE RELATIONS AND IDENTITIES FOR ORDER STATISTICS 2. 0. Introduction 5 2. 1. Relations for single moments 6 2. 2. Relations for product moments 9 2. 3. Relations for covariances 13 15 2. 4. Results for symmetric populations 2. 5. Results for normal population 17 20 2. 6. Results for two related populations 2. 7. Results for exchangeable variates 23 25 Exercises Chapter 3: BOUNDS ON EXPECTATIONS OF ORDER STATISTICS 3. 0. Introduction 38 3. 1. Universal bounds in the Li. d. case 38 3. 2. Variations on the Samuelson-Scott theme 43 3. 3.

The concept of conditional specification is not new. It is likely that earlier investigators in this area were deterred by computational difficulties encountered in the analysis of data following con­ ditionally specified models. Readily available computing power has swept away that roadblock. A broad spectrum of new flexible models may now be added to the researcher's tool box. This mono­ graph provides a preliminary guide to these models. Further development of inferential techniques, especially those involving concomitant variables, is clearly called for. We are grateful for invaluable assistance in the preparation of this monograph. In Riverside, Carole Arnold made needed changes in grammer and punctuation and Peggy Franklin miraculously transformed minute hieroglyphics into immaculate typescript. In Santander, Agustin Manrique ex­ pertly transformed rough sketches into clear diagrams. Finally, we thank the University of Cantabria for financial support which made possible Barry C. Arnold's enjoyable and productive visit to S- tander during the initial stages of the project. Barry C. Arnold Riverside, California USA Enrique Castillo Jose Maria Sarabia Santander, Cantabria Spain January, 1991 Contents 1 Conditional Specification 1 1.1 Why? ............. ........ . 1 1.2 How may one specify a bivariate distribution? 2 1.3 Early work on conditional specification 4 1.4 Organization of this monograph . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5 2 Basic Theorems 7 Compatible conditionals: The finite discrete case.

Efforts to visualize multivariate densities necessarily involve the use of cross-sections, or, equivalently, conditional densities. This book focuses on distributions that are completely specified in terms of conditional densities. They are appropriately used in any modeling situation where conditional information is completely or partially available. All statistical researchers seeking more flexible models than those provided by classical models will find conditionally specified distributions of interest.

The first and only comprehensive guide to modern record theory andits applications Although it is often thought of as a special topic in orderstatistics, records form a unique area, independent of the study ofsample extremes. Interest in records has increased steadily overthe years since Chandler formulated the theory of records in 1952.Numerous applications of them have been developed in such far-flungfields as meteorology, sports analysis, hydrology, and stock marketanalysis, to name just a few. And the literature on the subjectcurrently comprises papers and journal articles numbering in thehundreds. Which is why it is so nice to have this book devotedexclusively to this lively area of statistics. Written by an exceptionally well-qualified author team, Recordspresents a comprehensive treatment of record theory and itsapplications in a variety of disciplines. With the help of amultitude of fascinating examples, Professors Arnold, Balakrishnan,and Nagaraja help readers quickly master basic and advanced recordvalue concepts and procedures, from the classical record valuemodel to random and multivariate record models. The book follows arational textbook format, featuring witty and insightful chapterintroductions that help smooth transitions from one topic toanother and challenging chapter-end exercises, which expand on thematerial covered. An extensive bibliography and numerous referencesthroughout the text specify sources for further readings onrelevant topics. Records is a valuable professional resource forprobabilists and statisticians, in addition to appliedstatisticians, meteorologists, hydrologists, market analysts, andsports analysts. It also makes an excellent primary text forcourses in record theory and a supplement to order statisticscourses.

Enrique Castillo is a leading figure in several mathematical and engineering fields, having contributed seminal work in such areas as statistical modeling, extreme value analysis, multivariate distribution theory, Bayesian networks, neural networks, functional equations, artificial intelligence, linear algebra, optimization methods, numerical methods, reliability engineering, as well as sensitivity analysis and its applications. Organized to honor Castillo’s significant contributions, this volume is an outgrowth of the "International Conference on Mathematical and Statistical Modeling," and covers recent advances in the field.Applications to safety, reliability and life-testing, financial modeling, quality control, general inference, as well as neural networks and computational techniques are presented. The book is divided into nine major sections, which include distribution theory and applications, probability and statistics, order statistics and analysis, engineering modeling, extreme value theory, business and economics applications, statistical methods, applied mathematics, and discrete distributions.This comprehensive reference work will appeal to a diverse audience from the statistical, applied mathematics, engineering, and economics communities. Practitioners, researchers, and graduate students in mathematical and statistical modeling, optimization, and computing will benefit from this work.

Since the publication of the first edition over 30 years ago, the literature related to Pareto distributions has flourished to encompass computer-based inference methods. Pareto Distributions, Second Edition provides broad, up-to-date coverage of the Pareto model and its extensions. This edition expands several chapters to accommodate recent results and reflect the increased use of more computer-intensive inference procedures.New to the Second Edition New material on multivariate inequality Recent ways of handling the problems of inference for Pareto models and their generalizations and extensionsNew discussions of bivariate and multivariate income and survival modelsThis book continues to provide researchers with a useful resource for understanding the statistical aspects of Pareto and Pareto-like distributions. It covers income models and properties of Pareto distributions, measures of inequality for studying income distributions, inference procedures for Pareto distributions, and various multivariate Pareto distributions existing in the literature.

Efforts to visualize multivariate densities necessarily involve the use of cross-sections, or, equivalently, conditional densities. This book focuses on distributions that are completely specified in terms of conditional densities. They are appropriately used in any modeling situation where conditional information is completely or partially available. All statistical researchers seeking more flexible models than those provided by classical models will find conditionally specified distributions of interest.

This greatly expanded new edition includes recent research on stochastic, multivariate and group majorization, Lorenz order, and applications in physics and chemistry, in economics and political science, in matrix inequalities, and in probability and statistics.

This book was written to serve as a graduate-level textbook for special topics classes in mathematics, statistics, and economics, to introduce these topics to other researchers, and for use in short courses. It is an introduction to the theory of majorization and related notions, and contains detailed material on economic applications of majorization and the Lorenz order, investigating the theoretical aspects of these two interrelated orderings. Revising and expanding on an earlier monograph, Majorization and the Lorenz Order: A Brief Introduction, the authors provide a straightforward development and explanation of majorization concepts, addressing historical development of the topics, and providing up-to-date coverage of families of Lorenz curves. The exposition of multivariate Lorenz orderings sets it apart from existing treatments of these topics.Mathematicians, theoretical statisticians, economists, and other social scientists who already recognize the utility of the Lorenz order in income inequality contexts and arenas will find the book useful for its sound development of relevant concepts rigorously linked to both the majorization literature and the even more extensive body of research on economic applications. Barry C. Arnold, PhD, is Distinguished Professor in the Statistics Department at the University of California, Riverside. He is a Fellow of the American Statistical Society, the American Association for the Advancement of Science, and the Institute of Mathematical Statistics, and is an elected member of the International Statistical Institute. He is the author of more than two hundred publications and eight books.José María Sarabia, PhD, is Professor of Statistics and Quantitative Methods in Business and Economics in the Department of Economics at the University of Cantabria, Spain. He is author of more than one hundred and fifty publications and ten books and is an associate editor of several journals including TEST, Communications in Statistics, and Journal of Statistical Distributions and Applications.

In these cases, both the exact quantiles and the exact p-values of the likelihood ratio tests can be computed quickly and efficiently.The test statistics in question range from common ones, such as those used to test e.g.

Dedicated to one of the most outstanding researchers in the field of statistics, this volume in honor of C.R.

Dedicated to one of the most outstanding researchers in the field of statistics, this volume in honor of C.R.

This book of selected works exhibits a significant part of Carlos A. Coelho’s research activity in multivariate statistics, focusing on topics related to likelihood ratio tests. After two introductory chapters, including a biography and an overview of the contributions and related research, the articles are presented in their chronological order of publication, thus also showing the research path followed by the author. The volume shows how likelihood ratio tests useful in multivariate analysis can be developed, and how in most cases exact distributions can be obtained for their statistics in a manageable form. Furthermore, it shows how extremely sharp asymptotic (near-exact) approximations can be found in the remaining cases. Circular distributions are also addressed and, more generally, distributions of products of random variables. The book will be useful for advanced students as well as researchers and practitioners who deal with likelihood ratio tests in multivariate analysis.

This book was written to serve as a graduate-level textbook for special topics classes in mathematics, statistics, and economics, to introduce these topics to other researchers, and for use in short courses. It is an introduction to the theory of majorization and related notions, and contains detailed material on economic applications of majorization and the Lorenz order, investigating the theoretical aspects of these two interrelated orderings. Revising and expanding on an earlier monograph, Majorization and the Lorenz Order: A Brief Introduction, the authors provide a straightforward development and explanation of majorization concepts, addressing historical development of the topics, and providing up-to-date coverage of families of Lorenz curves. The exposition of multivariate Lorenz orderings sets it apart from existing treatments of these topics.Mathematicians, theoretical statisticians, economists, and other social scientists who already recognize the utility of the Lorenz order in income inequality contexts and arenas will find the book useful for its sound development of relevant concepts rigorously linked to both the majorization literature and the even more extensive body of research on economic applications. Barry C. Arnold, PhD, is Distinguished Professor in the Statistics Department at the University of California, Riverside. He is a Fellow of the American Statistical Society, the American Association for the Advancement of Science, and the Institute of Mathematical Statistics, and is an elected member of the International Statistical Institute. He is the author of more than two hundred publications and eight books.José María Sarabia, PhD, is Professor of Statistics and Quantitative Methods in Business and Economics in the Department of Economics at the University of Cantabria, Spain. He is author of more than one hundred and fifty publications and ten books and is an associate editor of several journals including TEST, Communications in Statistics, and Journal of Statistical Distributions and Applications.

In commemoration of the bicentennial of the birth of the “lady who gave the rose diagram to us”, this special contributed book pays a statistical tribute to Florence Nightingale. This book presents recent phenomenal developments, both in rigorous theory as well as in emerging methods, for applications in directional statistics, in 25 chapters with contributions from 65 renowned researchers from 25 countries. With the advent of modern techniques in statistical paradigms and statistical machine learning, directional statistics has become an indispensable tool. Ranging from data on circles to that on the spheres, tori and cylinders, this book includes solutions to problems on exploratory data analysis, probability distributions on manifolds, maximum entropy, directional regression analysis, spatio-directional time series, optimal inference, simulation, statistical machine learning with big data, and more, with their innovative applications to emerging real-life problems in astro-statistics, bioinformatics, crystallography, optimal transport, statistical process control, and so on.

In commemoration of the bicentennial of the birth of the “lady who gave the rose diagram to us”, this special contributed book pays a statistical tribute to Florence Nightingale. This book presents recent phenomenal developments, both in rigorous theory as well as in emerging methods, for applications in directional statistics, in 25 chapters with contributions from 65 renowned researchers from 25 countries. With the advent of modern techniques in statistical paradigms and statistical machine learning, directional statistics has become an indispensable tool. Ranging from data on circles to that on the spheres, tori and cylinders, this book includes solutions to problems on exploratory data analysis, probability distributions on manifolds, maximum entropy, directional regression analysis, spatio-directional time series, optimal inference, simulation, statistical machine learning with big data, and more, with their innovative applications to emerging real-life problems in astro-statistics, bioinformatics, crystallography, optimal transport, statistical process control, and so on.