Synopses & Reviews
The area of data analysis has been greatly affected by our computer age. For example, the issue of collecting and storing huge data sets has become quite simplified and has greatly affected such areas as finance and telecommunications. Even non-specialists try to analyze data sets and ask basic questions about their structure. One such question is whether one observes some type of invariance with respect to scale, a question that is closely related to the existence of long-range dependence in the data. This important topic of long-range dependence is the focus of this unique work, written by a number of specialists on the subject. The topics selected should give a good overview from the probabilistic and statistical perspective. Included will be articles on fractional Brownian motion, models, inequalities and limit theorems, periodic long-range dependence, parametric, semiparametric, and non-parametric estimation, long-memory stochastic volatility models, robust estimation, and prediction for long-range dependence sequences. For those graduate students and researchers who want to use the methodology and need to know the "tricks of the trade," there will be a special section called "Mathematical Techniques." Topics in the first part of the book are covered from probabilistic and statistical perspectives and include fractional Brownian motion, models, inequalities and limit theorems, periodic long-range dependence, parametric, semiparametric, and non-parametric estimation, long-memory stochastic volatility models, robust estimation, prediction for long-range dependence sequences. The reader is referred to more detailed proofs if already found in the literature. The last part of the book is devoted to applications in the areas of simulation, estimation and wavelet techniques, traffic in computer networks, econometry and finance, multifractal models, and hydrology. Diagrams and illustrations enhance the presentation. Each article begins with introductory background material and is accessible to mathematicians, a variety of practitioners, and graduate students. The work serves as a state-of-the art reference or graduate seminar text.
Review
"This book is a valuable and unique collection of material on LRD. It gathers together a large amount of material; the papers and their extensive references here give a broad and current summary of the research on the topic of long-range dependence." ---Journal of Statistical Physics "Long-range dependence concerns time-series models for phenomena in which correlations decay like a power law, thus much more slowly than in ARMA models.... The present volume contains 28 surveys of the field, treating and merging both theory and applications, giving a global picture of the field. The papers are grouped into four parts: I. Probabilistic Aspects; II. Statistical Aspects; III. Applications to Telecommunications, Economics, Hydrology, and Turbulence; IV. Practical Methodologies."--QUARTERLY OF APPLIED MATHEMATICS "The present volume contains 28 surveys of the field, treating and merging both theory and applications, giving a global picture of the field."(QUARTERLY OF APPLIED MATHEMATICS)
Synopsis
Time series research has been an area of considerable research activity over the past several decades. The essential ingredient --- the notion of time-dependence --- is required for measuring and then accurately predicting data to construct suitable models for diverse phenomena. This fairly self-contained volume, written by leading experts in their respective fields, especially focuses on the theoretical concepts, methodologies, and practical applications pertaining to self-similar processes and long-range dependent phenomena. Graduate students, researchers, and professionals in industry will benefit from the book.
Table of Contents
Preface * I. Probability * II. Statistics * III. Applications * IV. Methodology