The author has attempted to present a book that provides a non-technical introduction into the area of non-parametric density and regression function estimation. The application of these methods is discussed in terms of the S computing environment. Smoothing in high dimensions faces the problem of data sparseness. A principal feature of smoothing, the averaging of data points in a prescribed neighborhood, is not really practicable in dimensions greater than three if we have just one hundred data points. Additive models provide a way out of this dilemma; but, for their interactiveness and recursiveness, they require highly effective algorithms. For this purpose, the method of WARPing (Weighted Averaging using Rounded Points) is described in great detail.

This new edition includes the latest advances and developments in computational probability involving A Probability Programming Language (APPL). The book examines and presents, in a systematic manner, computational probability methods that encompass data structures and algorithms. The developed techniques address problems that require exact probability calculations, many of which have been considered intractable in the past. The book addresses the plight of the probabilist by providing algorithms to perform calculations associated with random variables. Computational Probability: Algorithms and Applications in the Mathematical Sciences, 2nd Edition begins with an introductory chapter that contains short examples involving the elementary use of APPL. Chapter 2 reviews the Maple data structures and functions necessary to implement APPL. This is followed by a discussion of the development of the data structures and algorithms (Chapters 3-6 for continuous random variables and Chapters 7-9 for discrete random variables) used in APPL. The book concludes with Chapters 10-15 introducing a sampling of various applications in the mathematical sciences. This book should appeal to researchers in the mathematical sciences with an interest in applied probability and instructors using the book for a special topics course in computational probability taught in a mathematics, statistics, operations research, management science, or industrial engineering department.

Main concepts of quasi-stationary distributions (QSDs) for killed processes are the focus of the present volume. For diffusions, the killing is at the boundary and for dynamical systems there is a trap. The authors present the QSDs as the ones that allow describing the long-term behavior conditioned to not being killed. Studies in this research area started with Kolmogorov and Yaglom and in the last few decades have received a great deal of attention. The authors provide the exponential distribution property of the killing time for QSDs, present the more general result on their existence and study the process of trajectories that survive forever. For birth-and-death chains and diffusions, the existence of a single or a continuum of QSDs is described. They study the convergence to the extremal QSD and give the classification of the survival process. In this monograph, the authors discuss Gibbs QSDs for symbolic systems and absolutely continuous QSDs for repellers. The findings described are relevant to researchers in the fields of Markov chains, diffusions, potential theory, dynamical systems, and in areas where extinction is a central concept. The theory is illustrated with numerous examples. The volume uniquely presents the distribution behavior of individuals who survive in a decaying population for a very long time. It also provides the background for applications in mathematical ecology, statistical physics, computer sciences, and economics.

Lazar Mayants is a recent Russian emigre noted for his work in theoretical physics. He was previously a professor at several universities of the Soviet Union and a distinguished member of the Academy of Sciences of the U.S.S.R, where he worked for about 30 years. In this book he presents a unique, extremely detailed, and embracive version of a subject that has suffered for a long time from numerous internal imperfections. His approach is new and original, the material covered features not only the foundations of the science of probability but also most of its applications, including statistical and quantum mechanics. The key methodolOgical principle underlying the book is of extraordinary significance and deserves special attention. The treatment excels in thoroughness of presentation, in its fulness of mathe matical detail and the abundance of physical examples. The book is intended for a wide range of people interested in probability and its connection with modern science. It is written as a text for advanced students, and I predict that a reader who masters all its contents will become an expert in the subject of both prob ability and its physical implications, while enjoying its understanding and use. HENRY MARGENAU Veritas nihil veretur nisi abscondi (truth 'What tremendously easy riddles you ask ' Humpty Dumpty growled out. fears nothing except being hidden). Latin proverb Lewis Carroll, Through the Looking Glass, Chap. 6. Preface The history of producing this book is rather complicated and not quite usual."

Over the last ten years the introduction of computer intensive statistical methods has opened new horizons concerning the probability models that can be fitted to genetic data, the scale of the problems that can be tackled and the nature of the questions that can be posed. In particular, the application of Bayesian and likelihood methods to statistical genetics has been facilitated enormously by these methods. Techniques generally referred to as Markov chain Monte Carlo (MCMC) have played a major role in this process, stimulating synergies among scientists in different fields, such as mathematicians, probabilists, statisticians, computer scientists and statistical geneticists. Specifically, the MCMC "revolution" has made a deep impact in quantitative genetics. This can be seen, for example, in the vast number of papers dealing with complex hierarchical models and models for detection of genes affecting quantitative or meristic traits in plants, animals and humans that have been published recently. This book, suitable for numerate biologists and for applied statisticians, provides the foundations of likelihood, Bayesian and MCMC methods in the context of genetic analysis of quantitative traits. Most students in biology and agriculture lack the formal background needed to learn these modern biometrical techniques. Although a number of excellent texts in these areas have become available in recent years, the basic ideas and tools are typically described in a technically demanding style, and have been written by and addressed to professional statisticians. For this reason, considerable more detail is offered than what may be warranted for a more mathematically apt audience. The book is divided into four parts. Part I gives a review of probability and distribution theory. Parts II and III present methods of inference and MCMC methods. Part IV discusses several models that can be applied in quantitative genetics, primarily from a Bayesian perspective. An effort has been made to relate biological to statistical parameters throughout, and examples are used profusely to motivate the developments. Daniel Sorensen is Research Leader in Biometrical Genetics, at the Department of Animal Breeding and Genetics in the Danish Institute of Agricultural Sciences. Daniel Gianola is Professor in the Animal Sciences, Biostatistics and Medical Informatics, and Dairy Science Departments of the University of Wisconsin-Madison. Gianola and Sorensen pioneered the introduction of Bayesian and MCMC methods in animal breeding. The authors have published and lectured extensively in applications of statistics to quantitative genetics.

The scope of this book is limited to heuristics, metaheuristics, and approximate methods and algorithms as applied to planning and scheduling problems. While it is not possible to give a comprehensive treatment of this topic in one book, the aim of this work is to provide the reader with a diverse set of planning and scheduling problems and different heuristic approaches to solve them. The problems range from traditional single stage and parallel machine problems to more modern settings such as robotic cells and flexible job shop networks. Furthermore, some chapters deal with deterministic problems while some others treat stochastic versions of the problems. Unlike most of the literature that deals with planning and scheduling problems in the manufacturing and production environments, in this book the environments were extended to nontraditional applications such as spatial scheduling (optimizing space over time), runway scheduling, and surgical scheduling. The solution methods used in the different chapters of the book also spread from well-established heuristics and metaheuristics such as Genetic Algorithms and Ant Colony Optimization to more recent ones such as Meta-RaPS.

This textbook provides a comprehensive introduction to probability and stochastic processes, and shows how these subjects may be applied in computer performance modeling. The author's aim is to derive probability theory in a way that highlights the complementary nature of its formal, intuitive, and applicative aspects while illustrating how the theory is applied in a variety of settings. Readers are assumed to be familiar with elementary linear algebra and calculus, including being conversant with limits, but otherwise, this book provides a self-contained approach suitable for graduate or advanced undergraduate students. The first half of the book covers the basic concepts of probability, including combinatorics, expectation, random variables, and fundamental theorems. In the second half of the book, the reader is introduced to stochastic processes. Subjects covered include renewal processes, queueing theory, Markov processes, matrix geometric techniques, reversibility, and networks of queues. Examples and applications are drawn from problems in computer performance modeling. Throughout, large numbers of exercises of varying degrees of difficulty will help to secure a reader's understanding of these important and fascinating subjects.

Introduction to the basic concepts of probability theory: independence, expectation, convergence in law and almost-sure convergence. Short expositions of more advanced topics such as Markov Chains, Stochastic Processes, Bayesian Decision Theory and Information Theory.

This book provides a panoramic view of theory and applications of Ageing and Dependence in the use of mathematical methods in reliability and survival analysis. Ageing and dependence are important characteristics in reliability and survival analysis. They affect decisions with regard to maintenance, repair/replacement, price setting, warranties, medical studies, and other areas. Most of the works containing the topics covered here are theoretical in nature. However, this book offers applications, exercises, and examples. It serves as a reference for professors and researchers involved in reliability and survival analysis.

VII Preface In many fields of mathematics, geometry has established itself as a fruitful method and common language for describing basic phenomena and problems as well as suggesting ways of solutions. Especially in pure mathematics this is ob vious and well-known (examples are the much discussed interplay between lin ear algebra and analytical geometry and several problems in multidimensional analysis). On the other hand, many specialists from applied mathematics seem to prefer more formal analytical and numerical methods and representations. Nevertheless, very often the internal development of disciplines from applied mathematics led to geometric models, and occasionally breakthroughs were b ed on geometric insights. An excellent example is the Klee-Minty cube, solving a problem of linear programming by transforming it into a geomet ric problem. Also the development of convex programming in recent decades demonstrated the power of methods that evolved within the field of convex geometry. The present book focuses on three applied disciplines: control theory, location science and computational geometry. It is our aim to demonstrate how methods and topics from convex geometry in a wider sense (separation theory of convex cones, Minkowski geometry, convex partitionings, etc.) can help to solve various problems from these disciplines."

Features: Second edition has been updated with a new chapter on Nonparametric Estimation; a significant update to the chapter on Statistical Decision Theory; and other updates throughout No requirement for heavy calculus, and simple questions throughout the text help students check their understanding of the material Each chapter also includes a set of exercises that range in level of difficulty Self-contained, and can be used by the students to understand the theory Chapters and sections marked by asterisks contain more advanced topics and may be omitted Special chapters on linear models and nonparametric statistics show how the main theoretical concepts can be applied to well-known and frequently used statistical tools

Mathematical Puzzle Tales from Mount Olympus uses fascinating tales from Greek Mythology as the background for introducing mathematics puzzles to the general public. A background in high school mathematics will be ample preparation for using this book, and it should appeal to anyone who enjoys puzzles and recreational mathematics. Features: Combines the arts and science, and emphasizes the fact that mathematics straddles both domains. Great resource for students preparing for mathematics competitions, and the trainers of such students.

The pioneering research of Hirotugu Akaike has an international reputation for profoundly affecting how data and time series are analyzed and modelled and is highly regarded by the statistical and technological communities of Japan and the world. His 1974 paper "A new look at the statistical model identification" (IEEE Trans Automatic Control, AC-19, 716-723) is one of the most frequently cited papers in the area of engineering, technology, and applied sciences (according to a 1981 Citation Classic of the Institute of Scientific Information). It introduced the broad scientific community to model identification using the methods of Akaike's criterion AIC. The AIC method is cited and applied in almost every area of physical and social science. The best way to learn about the seminal ideas of pioneering researchers is to read their original papers. This book reprints 29 papers of Akaike's more than 140 papers. This book of papers by Akaike is a tribute to his outstanding career and a service to provide students and researchers with access to Akaike's innovative and influential ideas and applications. To provide a commentary on the career of Akaike, the motivations of his ideas, and his many remarkable honors and prizes, this book reprints "A Conversation with Hirotugu Akaike" by David F. Findley and Emanuel Parzen, published in 1995 in the journal Statistical Science. This survey of Akaike's career provides each of us with a role model for how to have an impact on society by stimulating applied researchers to implement new statistical methods.

The book will serve primarily as a user's manual or desk reference for the expert witness-lawyer team and secondarily as a textbook or supplemental textbook for upper level undergraduate statistics students. It starts with two articles by masters of the trade, Paul Meier and Franklin Fisher. It then explains the distinction between the Frye and Daughbert standards for expert testimony, and how these standards play out in court. The bulk of the book is concerned with individual cases ranging over a wide variety of topics, such as electronic draw poker (does it require skill to play), employment discrimination (how to tell whether an employer discriminated against older workers in deciding whom to fire), driving while black (did the New Jersey State Police disproportionately stop blacks), jury representativeness (is a jury a representative cross section of the community), juries hearing death penalty cases (are such juries biased toward a guilty verdict, and does the Supreme Court care), the civil incarceration of violent sexual offenders after having served their jail sentences (can future dangerousness be predicted), do data from multiple choice examinations support an allegation of copying, whether rental agents in an apartment complex steered African-American prospects to one part of the complex, how much tax is owed after an audit that used a random sample, whether an inventor falsified his notebook in an effort to fool the Patent Office, and whether ballots had been tampered with in an election. The book concludes with two recent English cases, one in which a woman was accused of murdering her infant sons because both died of "cot death" or "sudden death syndrome", (she was convicted, but later exonerated), and how Bayesian analyses can (or more precisely), cannot be presented in UK courts. In each study, the statistical analysis is shaped to address the relevant legal questions, and draws on whatever methods in statistics might shed light on those questions.

This book provides an account for the non-specialist of the circle of ideas, results and techniques, which grew out in the study of Brownian motion and random obstacles. It also includes an overview of known results and connections with other areas of random media, taking a highly original and personal approach throughout.

The Equation of Knowledge: From Bayes' Rule to a Unified Philosophy of Science introduces readers to the Bayesian approach to science: teasing out the link between probability and knowledge. The author strives to make this book accessible to a very broad audience, suitable for professionals, students, and academics, as well as the enthusiastic amateur scientist/mathematician. This book also shows how Bayesianism sheds new light on nearly all areas of knowledge, from philosophy to mathematics, science and engineering, but also law, politics and everyday decision-making. Bayesian thinking is an important topic for research, which has seen dramatic progress in the recent years, and has a significant role to play in the understanding and development of AI and Machine Learning, among many other things. This book seeks to act as a tool for proselytising the benefits and limits of Bayesianism to a wider public. Features Presents the Bayesian approach as a unifying scientific method for a wide range of topics Suitable for a broad audience, including professionals, students, and academics Provides a more accessible, philosophical introduction to the subject that is offered elsewhere

The book concerns the notion of association in probability and statistics. Association and some other positive dependence notions were introduced in 1966 and 1967 but received little attention from the probabilistic and statistics community. The interest in these dependence notions increased in the last 15 to 20 years, and many asymptotic results were proved and improved. Despite this increased interest, characterizations and results remained essentially scattered in the literature published in different journals. The goal of this book is to bring together the bulk of these results, presenting the theory in a unified way, explaining relations and implications of the results. It will present basic definitions and characterizations, followed by a collection of relevant inequalities. These are then applied to characterize almost sure and weak convergence of sequences of associated variables. It will also cover applications of positive dependence to the characterization of asymptotic results in nonparametric statistics. The book is directed towards researchers in probability and statistics, with particular emphasis on people interested in nonparametric methods. It will also be of interest to graduate students in those areas. The book could also be used as a reference on association in a course covering dependent variables and their asymptotics.
As prerequisite, readers should have knowledge of basic probability on the reals and on metric spaces. Some acquaintance with the asymptotics of random functions, such us empirical processes and partial sums processes, is useful but not essential.

This book introduces the reader to the use of Monte Carlo methods for solving practical problems in radiation transport, and will also serve as a reference work for practitioners in the field. It assumes the reader has a general knowledge of calculus and radiation physics, and a knowledge of Fortran programming, but assumes no prior knowledge of stochastic methods or statistical physics. The subject is presented by a combination of theoretical development and practical calculations. Because Monte Carlo methods are closely linked to the use of computers, from the beginning the reader is taught to convert the theoretical constructs developed in the text into functional software for use on a personal computer. Example problems provide the reader with an in-depth understanding of the concepts presented and lead to the production of a unique learning tool, a probabilistic framework code that models in a simple manner the features of production of Monte Carlo transport codes. This framework code is developed in stages such that every function is understood, tested, and demonstrated - random sampling, generating random numbers, implementing geometric models, using variance reduction, tracking particles in a random walk, testing the thoroughness with which the problem phase space is sampled, scoring detectors, and obtaining estimates of uncertainty in results. Advanced topics covered include criticality, correlated sampling, adjoint transport, and neutron thermalization. Monte Carlo codes can produce highly precise wrong answers. The probability of this occurring is increased if production codes are run as opaque, black boxes' of software. This text attempts to make Monte Carlo into acomprehensible, usable tool for solving practical transport problems. It is suitable for advanced undergraduate and graduate students and researchers who wish to expand their knowledge of the Monte Carlo technique.

Semi-infinite programming (briefly: SIP) is an exciting part of mathematical programming. SIP problems include finitely many variables and, in contrast to finite optimization problems, infinitely many inequality constraints. Prob lems of this type naturally arise in approximation theory, optimal control, and at numerous engineering applications where the model contains at least one inequality constraint for each value of a parameter and the parameter, repre senting time, space, frequency etc., varies in a given domain. The treatment of such problems requires particular theoretical and numerical techniques. The theory in SIP as well as the number of numerical SIP methods and appli cations have expanded very fast during the last years. Therefore, the main goal of this monograph is to provide a collection of tutorial and survey type articles which represent a substantial part of the contemporary body of knowledge in SIP. We are glad that leading researchers have contributed to this volume and that their articles are covering a wide range of important topics in this subject. It is our hope that both experienced students and scientists will be well advised to consult this volume. We got the idea for this volume when we were organizing the semi-infinite pro gramming workshop which was held in Cottbus, Germany, in September 1996."

The current financial crisis has revealed serious flaws in models, measures and, potentially, theories, that failed to provide forward-looking expectations for upcoming losses originated from market risks. The Proceedings of the Perm Winter School 2011 propose insights on many key issues and advances in financial markets modeling and risk measurement aiming to bridge the gap. The key addressed topics include: hierarchical and ultrametric models of financial crashes, dynamic hedging, arbitrage free modeling the term structure of interest rates, agent based modeling of order flow, asset pricing in a fractional market, hedge funds performance and many more.

This volume covers recent developments in the design, operation, and management of mobile telecommunication and computer systems.

Uncertainty regarding loading and system parameters leads to challenging optimization and robustness issues. Stochastic modeling combined with optimization theory ensures the optimum end-to-end performance of telecommunication or computer network systems. In view of the diverse design options possible, supporting models have many adjustable parameters and choosing the best set for a particular performance objective is delicate and time-consuming. An optimization based approach determines the optimal possible allocation for these parameters.

Researchers and graduate students working at the interface of telecommunications and operations research will benefit from this book. Due to the practical approach, this book will also serve as a reference tool for scientists and engineers in telecommunication and computer networks who depend upon optimization.

Detailed lecture notes on six topics at the forefront of current research in numerical analysis and applied mathematics, with each set of notes presenting a self-contained guide to a current research area and supplemented by an extensive bibliography. In addition, most of the notes contain detailed proofs of the key results. They start from a level suitable for first year graduates in applied mathematics, mathematical analysis or numerical analysis, and proceed to current research topics. Readers will thus quickly gain an insight into the important results and techniques in each area without recourse to the large research literature. Current (unsolved) problems are also described, and directions for future research given.

This textbook provides a broad overview of the basic theory and methods of applied multivariate analysis. The presentation integrates both theory and practice including both the analysis of formal linear multivariate models and exploratory data analysis techniques. Each chapter contains the development of basic theoretical results with numerous applications illustrated using examples from the social and behavioral sciences, and other disciplines. All examples are analyzed using SAS for Windows Version 8.0. The book includes an overview of vectors, matrices, multivariate distribution theory, and multivariate linear models. Topics discussed include multivariate regression, multivariate analysis of variance for fixed and mixed models, seemingly unrelated regression models and repeated measurement models. While standard procedures for estimating model parameters and testing multivariate hypotheses, as well as simultaneous test procedures, are discussed and illustrated in the text, the text also includes tests of multivariate normality with chi-square and beta plots, tests of multivariate nonadditivity, tests of covariance structure, tests of nonnested hypotheses, and the assessment of model assumptions. Other topics illustrated in the text include discriminant and classification analysis, principal component analysis, canonical correlation analysis, exploratory factor analysis, cluster analysis, multidimension scaling, and structural equation modeling. The text should appeal to practitioners, researchers, and applied statisticians. It may be used in a one-semester course in applied multivariate analysis for practitioners and researchers, or as a two-semester course for majors in applied statistics. Because most data analyzed in the social and behavioral sciences and other disciplines involve many continuous variables, the techniques and examples. SAS programs for this book are available on the Springer website.

Volume 1: (edited by Keith W. Hipel) In this landmark collection of papers, highly respected scientists and engineers from around the world present the latest research results in extreme value analyses for floods and droughts. Two approaches that are commonly employed in flood frequency analyses are the maximum annual flood and partial duration series or peak over threshold procedures. Recent theoretical advances as well as illustrative applications are described in detail for each of these approaches. Additionally, droughts and storms are systematically studied using appropriate probabilistic models. A major part of the volume is devoted to frequency analyses and fitting extreme value distributions to hydrological data. Other thought-provoking topics include regionalization techniques, distributed models, entropy and fractal analysis. Volume 1 is of interest to researchers, teachers, students and practitioners who wish to place themselves at the leading edge of flood frequency and drought analyses. Volume 2: (edited by Keith W. Hipel) World renowned scientists present valuable contributions to stochastic and statistical modelling of groundwater and surface water systems. The philosophy of probabilistic modelling in the hydrological sciences is put into proper perspective and the importance of stochastic differential equations in the environmental sciences is explained and illustrated. The new research ideas put forward in groundwater modelling will assist decision makers in tackling challenging problems such as controlling pollution of underground aquifers and obtaining adequate water supplies. Additionally, different types of stochastic models are used in modelling a range ofinteresting surface water problems. Other topics covered in this landmark volume include stochastic optimization, moment analysis, carbon dioxide modelling and rainfall prediction. Volume 2 is of interest to researchers, teachers, students and practitioners who wish to be at the leading edge of stochastic and statistical modelling in the environmental sciences. Volume 3: (edited by Keith W. Hipel; A. Ian McLeod; U.S. Panu; Vijay P. Singh) International experts from around the globe present a rich variety of intriguing developments in time series analysis in hydrology and environmental engineering. Climatic change is of great concern to everyone and significant contributions to this challenging research topic are put forward by internationally renowned authors. A range of interesting applications in hydrological forecasting are given for case studies in reservoir operation in North America, Asia and South America. Additionally, progress in entropy research is described and entropy concepts are applied to various water resource systems problems. Neural networks are employed for forecasting runoff and water demand. Moreover, graphical, nonparametric and parametric trend analyses methods are compared and applied to water quality time series. Other topics covered in this landmark volume include spatial analyses, spectral analyses and different methods for stream-flow modelling. Volume 3 constitutes an invaluable resource for researchers, teachers, students and practitioners who wish to be at the forefront of time series analysis in the environmental sciences. Volume 4: (edited by Keith W. Hipel; Liping Fang) In this landmark set of papers, experts from around the world present the latest andmost promising approaches to both the theory and practice of effective environmental management. To achieve sustainable development, organizations and individual citizens must comply with environmental laws and regulations. Accordingly, a major contribution of this book is the presentation of original techniques for designing effective environmental policies, regulations, inspection procedures and monitoring systems. Interesting methods for modelling risk and decision making problems are discussed from an environmental management perspective. Moreover, knowledge-based techniques for handling environmental problems are also investigated. Finally, the last main part of the book describes optimal approaches to reservoir operation and control that take into account appropriate multiple objectives. Volume 4 is of direct interest to researchers, teachers, students and practitioners concerned with the latest developments in environmental management and sustainable development.