This volume presents some of the most influential papers published by Rabi N. Bhattacharya, along with commentaries from international experts, demonstrating his knowledge, insight, and influence in the field of probability and its applications. For more than three decades, Bhattacharya has made significant contributions in areas ranging from theoretical statistics via analytical probability theory, Markov processes, and random dynamics to applied topics in statistics, economics, and geophysics. Selected reprints of Bhattacharya's papers are divided into three sections: Modes of Approximation, Large Times for Markov Processes, and Stochastic Foundations in Applied Sciences. The accompanying articles by the contributing authors not only help to position his work in the context of other achievements, but also provide a unique assessment of the state of their individual fields, both historically and for the next generation of researchers. Rabi N. Bhattacharya: Selected Papers will be a valuable resource for young researchers entering the diverse areas of study to which Bhattacharya has contributed. Established researchers will also appreciate this work as an account of both past and present developments and challenges for the future.

These notes are based on lectures presented during the seminar on " Asymptotic Statistics" held at SchloB Reisensburg, Gunzburg, May 29-June 5, 1988. They consist of two parts, the theory of asymptotic expansions in statistics and probabilistic aspects of the asymptotic distribution theory in nonparametric statistics. Our intention is to provide a comprehensive presentation of these two subjects, leading from elementary facts to the advanced theory and recent results. Prospects for further research are also included. We would like to thank all participants for their stimulating discussions and their interest in the subjects, which made lecturing very pleasant. Special thanks are due H. Zimmer for her excellent typing. We would also like to take this opportunity to to express our thanks to the Gesellschaft fur mathematische Forschung and to the Deutsche Mathematiker Vereinigung, especially to Professor G. Fischer, for the opportunity to present these lectures and to the Birkhauser Verlag for the publication of these lecture notes. R. Bhattacharya, M. Denker Part I: Asymptotic Expansions in Statistics Rabi Bhattacharya 11 1. CRAMER-EDGEWORTH EXPANSIONS Let Q be a probability measure on (IRk, B"), B" denoting the Borel sigmafield on IR". Assume that the s - th absolute moment of Q is finite, (1.1) P. := J II x lis Q(dx) < 00, for some integer s;::: 3, and that Q is normalized, (1.2) J x(i)Q(dx) = 0 (1 ~ i ~ k), J x(i)x(j)Q(dx) = Dij (1 ~ i,j ~ k).

The present book is based on a course developed as partofthe large NSF-funded GatewayCoalitionInitiativeinEngineeringEducationwhichincludedCaseWest ern Reserve University, Columbia University, Cooper Union, Drexel University, Florida International University, New Jersey Institute ofTechnology, Ohio State University, University ofPennsylvania, Polytechnic University, and Universityof South Carolina. The Coalition aimed to restructure the engineering curriculum by incorporating the latest technological innovations and tried to attract more and betterstudents to engineering and science. Draftsofthis textbookhave been used since 1992instatisticscoursestaughtatCWRU, IndianaUniversity, Bloomington, and at the universities in Gottingen, Germany, and Grenoble, France. Another purpose of this project was to develop a courseware that would take advantage ofthe Electronic Learning Environment created by CWRUnet-the all fiber-optic Case Western Reserve University computer network, and its ability to let students run Mathematica experiments and projects in their dormitory rooms, and interactpaperlessly with the instructor. Theoretically, onecould try togothroughthisbook withoutdoing Mathematica experimentsonthecomputer, butitwouldbelikeplayingChopin's Piano Concerto in E-minor, or Pink Floyd's The Wall, on an accordion. One would get an idea ofwhatthe tune was without everexperiencing the full richness andpowerofthe entire composition, and the whole ambience would be miscued."

This volume presents some of the most influential papers published by Rabi N. Bhattacharya, along with commentaries from international experts, demonstrating his knowledge, insight, and influence in the field of probability and its applications. For more than three decades, Bhattacharya has made significant contributions in areas ranging from theoretical statistics via analytical probability theory, Markov processes, and random dynamics to applied topics in statistics, economics, and geophysics. Selected reprints of Bhattacharya's papers are divided into three sections: Modes of Approximation, Large Times for Markov Processes, and Stochastic Foundations in Applied Sciences. The accompanying articles by the contributing authors not only help to position his work in the context of other achievements, but also provide a unique assessment of the state of their individual fields, both historically and for the next generation of researchers. Rabi N. Bhattacharya: Selected Papers will be a valuable resource for young researchers entering the diverse areas of study to which Bhattacharya has contributed. Established researchers will also appreciate this work as an account of both past and present developments and challenges for the future.

This textbook integrates traditional statistical data analysis with new computational experimentation capabilities and concepts of algorithmic complexity and chaotic behavior in nonlinear dynamic systems. This was the first advanced text/reference to bring together such a comprehensive variety of tools for the study of random phenomena occurring in engineering and the natural, life, and social sciences. The crucial computer experiments are conducted using the readily available computer program Mathematica (R) Uncertain Virtual Worlds (TM) software packages which optimize and facilitate the simulation environment. Brief tutorials are included that explain how to use the Mathematica (R) programs for effective simulation and computer experiments. Large and original real-life data sets are introduced and analyzed as a model for independent study. This is an excellent classroom tool and self-study guide. The material is presented in a clear and accessible style providing numerous exercises and bibliographical notes suggesting further reading. Topics and Features Comprehensive and integrated treatment of uncertainty arising in engineering and scientific phenomena - algorithmic complexity, statistical independence, and nonlinear chaotic behavior Extensive exercise sets, examples, and Mathematica (R) computer experiments that reinforce concepts and algorithmic methods Thorough presentation of methods of data compression and representation Algorithmic approach to model selection and design of experiments Large data sets and 13 Mathematica (R)-based Uncertain Virtual Worlds (TM) programs and code This text is an excellent resource for all applied statisticians, engineers, and scientists who need to use modern statistical analysis methods to investigate and model their data. The present, softcover reprint is designed to make this classic textbook available to a wider audience.

The present book is based on a course developed as partofthe large NSF-funded GatewayCoalitionInitiativeinEngineeringEducationwhichincludedCaseWest ern Reserve University, Columbia University, Cooper Union, Drexel University, Florida International University, New Jersey Institute ofTechnology, Ohio State University, University ofPennsylvania, Polytechnic University, and Universityof South Carolina. The Coalition aimed to restructure the engineering curriculum by incorporating the latest technological innovations and tried to attract more and betterstudents to engineering and science. Draftsofthis textbookhave been used since 1992instatisticscoursestaughtatCWRU, IndianaUniversity, Bloomington, and at the universities in Gottingen, Germany, and Grenoble, France. Another purpose of this project was to develop a courseware that would take advantage ofthe Electronic Learning Environment created by CWRUnet-the all fiber-optic Case Western Reserve University computer network, and its ability to let students run Mathematica experiments and projects in their dormitory rooms, and interactpaperlessly with the instructor. Theoretically, onecould try togothroughthisbook withoutdoing Mathematica experimentsonthecomputer, butitwouldbelikeplayingChopin's Piano Concerto in E-minor, or Pink Floyd's The Wall, on an accordion. One would get an idea ofwhatthe tune was without everexperiencing the full richness andpowerofthe entire composition, and the whole ambience would be miscued."

These notes are based on lectures presented during the seminar on " Asymptotic Statistics" held at SchloB Reisensburg, Gunzburg, May 29-June 5, 1988. They consist of two parts, the theory of asymptotic expansions in statistics and probabilistic aspects of the asymptotic distribution theory in nonparametric statistics. Our intention is to provide a comprehensive presentation of these two subjects, leading from elementary facts to the advanced theory and recent results. Prospects for further research are also included. We would like to thank all participants for their stimulating discussions and their interest in the subjects, which made lecturing very pleasant. Special thanks are due H. Zimmer for her excellent typing. We would also like to take this opportunity to to express our thanks to the Gesellschaft fur mathematische Forschung and to the Deutsche Mathematiker Vereinigung, especially to Professor G. Fischer, for the opportunity to present these lectures and to the Birkhauser Verlag for the publication of these lecture notes. R. Bhattacharya, M. Denker Part I: Asymptotic Expansions in Statistics Rabi Bhattacharya 11 1. CRAMER-EDGEWORTH EXPANSIONS Let Q be a probability measure on (IRk, B"), B" denoting the Borel sigmafield on IR". Assume that the s - th absolute moment of Q is finite, (1.1) P. := J II x lis Q(dx) < 00, for some integer s;::: 3, and that Q is normalized, (1.2) J x(i)Q(dx) = 0 (1 ~ i ~ k), J x(i)x(j)Q(dx) = Dij (1 ~ i,j ~ k).

Dynamische Systeme stellen einen unverzichtbaren Bestandteil mathematischer
Modellbildung fur Anwendungen aller Art dar, angefangen von Physik uber Biologie
bis hin zur Informatik. Dieser Band fuhrt in diese Theorie ein und beschreibt Methoden und Dynamiken, wie sie fur eine systematische Modellbildung auch in den Anwendungen notwendig erscheinen. Wesentliche Grundzuge der Theorie werden beispielhaft im ersten Kapitel erlautert. Es schliesst sich eine Einfuhrung in niedrig-dimensionale Dynamiken an (u.a. rationale Funktionen), gefolgt von topologischer Dynamik (z.B. Attraktoren, Entropie und chaotisches Verhalten), differenzierbarer Dynamik (z.B. Liapunoff-Exponenten, Strukturstabilitat und Hyperbolizitat), Ergodentheorie (z.B. Ergodensatze, invariante Masse, Konservativitat) und schliesslich thermodynamischer Formalismus (z.B. Gibbs-Theorie, Zetafunktionen)."