This volume presents the published Proceedings of the joint meeting of GUM92 and the 7th International Workshop on Statistical Modelling, held in Munich, Germany from 13 to 17 July 1992. The meeting aimed to bring together researchers interested in the development and applications of generalized linear modelling in GUM and those interested in statistical modelling in its widest sense. This joint meeting built upon the success of previous workshops and GUM conferences. Previous GUM conferences were held in London and Lancaster, and a joint GUM Conference/4th Modelling Workshop was held in Trento. (The Proceedings of previous GUM conferences/Statistical Modelling Workshops are available as numbers 14 , 32 and 57 of the Springer Verlag series of Lecture Notes in Statistics). Workshops have been organized in Innsbruck, Perugia, Vienna, Toulouse and Utrecht. (Proceedings of the Toulouse Workshop appear as numbers 3 and 4 of volume 13 of the journal Computational Statistics and Data Analysis). Much statistical modelling is carried out using GUM, as is apparent from many of the papers in these Proceedings. Thus the Programme Committee were also keen on encouraging papers which addressed problems which are not only of practical importance but which are also relevant to GUM or other software development. The Programme Committee requested both theoretical and applied papers. Thus there are papers in a wide range of practical areas, such as ecology, breast cancer remission and diabetes mortality, banking and insurance, quality control, social mobility, organizational behaviour.

The analysis of data by statistical modelling is becoming increasingly important. This book presents both the theory of statistical modelling with generalized linear models and the application of the theory to practical problems using the widely available package GLIM. The authors have takenpains to integrate the theory with many practical examples which illustrate the value of interactive statistical modelling. Throughout the book theoretical issues of formulating and simplifying models are discussed, as are problems of validating the models by the detection of outliers and influential observations. The book arises from short courses given at the University of Lancaster's Centre for Applied Statistics, with an emphasis on practical programming in GLIM and numerous examples. A wide range of case studies is provided, using the normal, binomial, Poisson, multinomial, gamma, exponential andWeibull distributions. A feature of the book is a detailed discussion of survival analysis. Statisticians working in a wide range of fields, including biomedical and social sciences, will find this book an invaluable desktop companion to aid their statistical modelling. It will also provide a text for students meeting the ideas of statistical modelling for the first time.

This volume presents the published proceedings of the lOth International Workshop on Statistical Modelling, to be held in Innsbruck, Austria from 10 to 14 July, 1995. This workshop marks an important anniversary. The inaugural workshop in this series also took place in Innsbruck in 1986, and brought together a small but enthusiastic group of thirty European statisticians interested in statistical modelling. The workshop arose out of two G LIM conferences in the U. K. in London (1982) and Lancaster (1985), and from a num ber of short courses organised by Murray Aitkin and held at Lancaster in the early 1980s, which attracted many European statisticians interested in Generalised Linear Modelling. The inaugural workshop in Innsbruck con centrated on GLMs and was characterised by a number of features - a friendly and supportive academic atmosphere, tutorial sessions and invited speakers presenting new developments in statistical modelling, and a very well organised social programme. The academic programme allowed plenty of time for presentation and for discussion, and made available copies of all papers beforehand. Over the intervening years, the workshop has grown substantially, and now regularly attracts over 150 participants. The scope of the workshop is now much broader, reflecting the growth in the subject of statistical modelling over ten years. The elements ofthe first workshop, however, are still present, and participants always find the meetings relevant and stimulating.

Monte Carlo methods are numerical methods based on random sampling and quasi-Monte Carlo methods are their deterministic versions. This volume contains the refereed proceedings of the Second International Conference on Monte Carlo and Quasi-Monte Carlo Methods in Scientific Computing which was held at the University of Salzburg (Austria) from July 9--12, 1996. The conference was a forum for recent progress in the theory and the applications of these methods. The topics covered in this volume range from theoretical issues in Monte Carlo and simulation methods, low-discrepancy point sets and sequences, lattice rules, and pseudorandom number generation to applications such as numerical integration, numerical linear algebra, integral equations, binary search, global optimization, computational physics, mathematical finance, and computer graphics. These proceedings will be of interest to graduate students and researchers in Monte Carlo and quasi-Monte Carlo methods, to numerical analysts, and to practitioners of simulation methods.

This volume is a selection of papers presented at the Fourth International Workshop on Artificial Intelligence and Statistics held in January 1993. These biennial workshops have succeeded in bringing together researchers from Artificial Intelligence and from Statistics to discuss problems of mutual interest. The exchange has broadened research in both fields and has strongly encour aged interdisciplinary work. The theme ofthe 1993 AI and Statistics workshop was: "Selecting Models from Data". The papers in this volume attest to the diversity of approaches to model selection and to the ubiquity of the problem. Both statistics and artificial intelligence have independently developed approaches to model selection and the corresponding algorithms to implement them. But as these papers make clear, there is a high degree of overlap between the different approaches. In particular, there is agreement that the fundamental problem is the avoidence of "overfitting"-Le., where a model fits the given data very closely, but is a poor predictor for new data; in other words, the model has partly fitted the "noise" in the original data.

Correlated data arise in numerous contexts across a wide spectrum of subject-matter disciplines. Modeling such data present special challenges and opportunities that have received increasing scrutiny by the statistical community in recent years. In October 1996 a group of 210 statisticians and other scientists assembled on the small island of Nantucket, U. S. A. , to present and discuss new developments relating to Modelling Longitudinal and Spatially Correlated Data: Methods, Applications, and Future Direc tions. Its purpose was to provide a cross-disciplinary forum to explore the commonalities and meaningful differences in the source and treatment of such data. This volume is a compilation of some of the important invited and volunteered presentations made during that conference. The three days and evenings of oral and displayed presentations were arranged into six broad thematic areas. The session themes, the invited speakers and the topics they addressed were as follows: • Generalized Linear Models: Peter McCullagh-"Residual Likelihood in Linear and Generalized Linear Models" • Longitudinal Data Analysis: Nan Laird-"Using the General Linear Mixed Model to Analyze Unbalanced Repeated Measures and Longi tudinal Data" • Spatio---Temporal Processes: David R. Brillinger-"Statistical Analy sis of the Tracks of Moving Particles" • Spatial Data Analysis: Noel A. Cressie-"Statistical Models for Lat tice Data" • Modelling Messy Data: Raymond J. Carroll-"Some Results on Gen eralized Linear Mixed Models with Measurement Error in Covariates" • Future Directions: Peter J.

The book deals with bilinear forms in real random vectors and their generalizations as well as zonal polynomials and their applications in handling generalized quadratic and bilinear forms. The book is mostly self-contained. It starts from basic principles and brings the readers to the current research level in these areas. It is developed with detailed proofs and illustrative examples for easy readability and self-study. Several exercises are proposed at the end of the chapters. The complicated topic of zonal polynomials is explained in detail in this book. The book concentrates on the theoretical developments in all the topics covered. Some applications are pointed out but no detailed application to any particular field is attempted. This book can be used as a textbook for a one-semester graduate course on quadratic and bilinear forms and/or on zonal polynomials. It is hoped that this book will be a valuable reference source for graduate students and research workers in the areas of mathematical statistics, quadratic and bilinear forms and their generalizations, zonal polynomials, invariant polynomials and related topics, and will benefit statisticians, mathematicians and other theoretical and applied scientists who use any of the above topics in their areas. Chapter 1 gives the preliminaries needed in later chapters, including some Jacobians of matrix transformations. Chapter 2 is devoted to bilinear forms in Gaussian real ran dom vectors, their properties, and techniques specially developed to deal with bilinear forms where the standard methods for handling quadratic forms become complicated.

The exponential increase in the use of MCMC methods and the corre sponding applications in domains of even higher complexity have caused a growing concern about the available convergence assessment methods and the realization that some of these methods were not reliable enough for all-purpose analyses. Some researchers have mainly focussed on the con vergence to stationarity and the estimation of rates of convergence, in rela tion with the eigenvalues of the transition kernel. This monograph adopts a different perspective by developing (supposedly) practical devices to assess the mixing behaviour of the chain under study and, more particularly, it proposes methods based on finite (state space) Markov chains which are obtained either through a discretization of the original Markov chain or through a duality principle relating a continuous state space Markov chain to another finite Markov chain, as in missing data or latent variable models. The motivation for the choice of finite state spaces is that, although the resulting control is cruder, in the sense that it can often monitor con vergence for the discretized version alone, it is also much stricter than alternative methods, since the tools available for finite Markov chains are universal and the resulting transition matrix can be estimated more accu rately. Moreover, while some setups impose a fixed finite state space, other allow for possible refinements in the discretization level and for consecutive improvements in the convergence monitoring.

In the present work I want to show a mathematical study of the statistical notion of sufficiency mainly for undominated statistical experiments. The famous Burkholder's (1961) and Pitcher's(1957) examples motivated some researchers to develop new theory of sufficiency. Le Cam (1964) is probably the most excellent paper in this field of study. This note also belongs to the same area. Though it is more restrictive than Le Cam's paper(1964), a study which is connected more directly with the classical papers of Halmos and Savage(1949) , and Bahadur(1954) is shown. Namely I want to develop a study based on the notion of pivotal measure which was introduced by Halmos and Savage(1949) . It is great pleasure to have this opportunity to thank Professor H. Heyer and Professor H. Morimoto for their careful reading the manuscript and valuable comments on it. I am also thankful to Professor H. Luschgy and Professor D. Mussmann for thei r proposal of wr i ting "the note". I would like to dedicate this note to the memory of my father Eizo.

The primary objective of this book is to provide an introduction to the econometric modeling of count data for graduate students and researchers. It should serve anyone whose interest lies either in developing the field fur ther, or in applying existing methods to empirical questions. Much of the material included in this book is not specific to economics, or to quantita tive social sciences more generally, but rather extends to disciplines such as biometrics and technometrics. Applications are as diverse as the number of congressional budget vetoes, the number of children in a household, and the number of mechanical defects in a production line. The unifying theme is a focus on regression models in which a dependent count variable is modeled as a function of independent variables which mayor may not be counts as well. The modeling of count data has come of age. Inclusion of some of the fundamental models in basic textbooks, and implementation on standard computer software programs bear witness to that. Based on the standard Poisson regression model, numerous extensions and alternatives have been developed to address the common challenges faced in empirical modeling (unobserved heterogeneity, selectivity, endogeneity, measurement error, and dependent observations in the context of panel data or multivariate data, to name but a few) as well as the challenges that are specific to count data (e. g. , over dispersion and underdispersion).