Anthony C Atkinson – författare

Randomised Response-Adaptive Designs in Clinical Trials presents methods for the randomised allocation of treatments to patients in sequential clinical trials. Emphasizing the practical application of clinical trial designs, the book is designed for medical and applied statisticians, clinicians, and statisticians in training. After introducing clinical trials in drug development, the authors assess a simple adaptive design for binary responses without covariates. They discuss randomisation and covariate balance in normally distributed responses and cover many important response-adaptive designs for binary responses. The book then develops response-adaptive designs for continuous and longitudinal responses, optimum designs with covariates, and response-adaptive designs with covariates. It also covers response-adaptive designs that are derived by optimising an objective function subject to constraints on the variance of estimated parametric functions. The concluding chapter explores future directions in the development of adaptive designs.

Why We Wrote This Book This book is about using graphs to explore and model continuous multi­ variate data. Such data are often modelled using the multivariate normal distribution and, indeed, there is a literatme of weighty statistical tomes presenting the mathematical theory of this activity. Our book is very dif­ ferent. Although we use the methods described in these books, we focus on ways of exploring whether the data do indeed have a normal distribution. We emphasize outlier detection, transformations to normality and the de­ tection of clusters and unsuspected influential subsets. We then quantify the effect of these departures from normality on procedures such as dis­ crimination and duster analysis. The normal distribution is central to our book because, subject to our exploration of departures, it provides useful models for many sets of data. However, the standard estimates of the parameters, especially the covari­ ance matrix of the observations, are highly sensitive to the presence of outliers. This is both a blessing and a curse. It is a blessing because, if we estimate the parameters with the outliers excluded, their effect is appre­ ciable and apparent if we then include them for estimation. It is however a curse because it can be hard to detect which observations are outliers. We use the forward search for this purpose.

Why We Wrote This Book This book is about using graphs to explore and model continuous multi­ variate data. Such data are often modelled using the multivariate normal distribution and, indeed, there is a literatme of weighty statistical tomes presenting the mathematical theory of this activity. Our book is very dif­ ferent. Although we use the methods described in these books, we focus on ways of exploring whether the data do indeed have a normal distribution. We emphasize outlier detection, transformations to normality and the de­ tection of clusters and unsuspected influential subsets. We then quantify the effect of these departures from normality on procedures such as dis­ crimination and duster analysis. The normal distribution is central to our book because, subject to our exploration of departures, it provides useful models for many sets of data. However, the standard estimates of the parameters, especially the covari­ ance matrix of the observations, are highly sensitive to the presence of outliers. This is both a blessing and a curse. It is a blessing because, if we estimate the parameters with the outliers excluded, their effect is appre­ ciable and apparent if we then include them for estimation. It is however a curse because it can be hard to detect which observations are outliers. We use the forward search for this purpose.

Randomised Response-Adaptive Designs in Clinical Trials presents methods for the randomised allocation of treatments to patients in sequential clinical trials. Emphasizing the practical application of clinical trial designs, the book is designed for medical and applied statisticians, clinicians, and statisticians in training. After introducing clin

Why We Wrote This Book This book is about using graphs to explore and model continuous multi­ variate data. Such data are often modelled using the multivariate normal distribution and, indeed, there is a literatme of weighty statistical tomes presenting the mathematical theory of this activity. Our book is very dif­ ferent. Although we use the methods described in these books, we focus on ways of exploring whether the data do indeed have a normal distribution. We emphasize outlier detection, transformations to normality and the de­ tection of clusters and unsuspected influential subsets. We then quantify the effect of these departures from normality on procedures such as dis­ crimination and duster analysis. The normal distribution is central to our book because, subject to our exploration of departures, it provides useful models for many sets of data. However, the standard estimates of the parameters, especially the covari­ ance matrix of the observations, are highly sensitive to the presence of outliers. This is both a blessing and a curse. It is a blessing because, if we estimate the parameters with the outliers excluded, their effect is appre­ ciable and apparent if we then include them for estimation. It is however a curse because it can be hard to detect which observations are outliers. We use the forward search for this purpose.

The International Statistical Institute was founded in 1885 and is therefore one of the world''s oldest international scientific societies. The field of statistics is still expanding rapidly and possesses a rich variety of applications in many areas of human activity such as science, government, business, industry, and everyday affairs. In consequence, the celebration of the Institute''s centenary in 1985 is of considerable interest not only to statisticians but also more widely to the international scientific community. As part of its centennial celebration planning the Institute decided to publish a volume of papers representing the immensely wide range of interests encompassed by statistics in its international context, viewed both from a historical and from a contemporary standpoint. We were fortunate in securing the services of Anthony Atkinson and Stephen Fienberg as Editors of this volume: they have worked hard over a period of several years to put together a most fascinating collection of papers. On behalf of the Institute it is my pleasant duty to thank them and the authors for their contributions. J. DURBIN, President International Statistical Institute Preface The papers in this volume were prepared to help celebrate the centenary of the International Statistical Institute. During the lSI''s first 100 years statistics has matured, both as a scientific discipline and as a profession, in ways that the lSI''s founders could not possibly have imagined.

The International Statistical Institute was founded in 1885 and is therefore one of the world's oldest international scientific societies. The field of statistics is still expanding rapidly and possesses a rich variety of applications in many areas of human activity such as science, government, business, industry, and everyday affairs. In consequence, the celebration of the Institute's centenary in 1985 is of considerable interest not only to statisticians but also more widely to the international scientific community. As part of its centennial celebration planning the Institute decided to publish a volume of papers representing the immensely wide range of interests encompassed by statistics in its international context, viewed both from a historical and from a contemporary standpoint. We were fortunate in securing the services of Anthony Atkinson and Stephen Fienberg as Editors of this volume: they have worked hard over a period of several years to put together a most fascinating collection of papers. On behalf of the Institute it is my pleasant duty to thank them and the authors for their contributions. J. DURBIN, President International Statistical Institute Preface The papers in this volume were prepared to help celebrate the centenary of the International Statistical Institute. During the lSI's first 100 years statistics has matured, both as a scientific discipline and as a profession, in ways that the lSI's founders could not possibly have imagined.

Randomised Response-Adaptive Designs in Clinical Trials presents methods for the randomised allocation of treatments to patients in sequential clinical trials. Emphasizing the practical application of clinical trial designs, the book is designed for medical and applied statisticians, clinicians, and statisticians in training. After introducing clinical trials in drug development, the authors assess a simple adaptive design for binary responses without covariates. They discuss randomisation and covariate balance in normally distributed responses and cover many important response-adaptive designs for binary responses. The book then develops response-adaptive designs for continuous and longitudinal responses, optimum designs with covariates, and response-adaptive designs with covariates. It also covers response-adaptive designs that are derived by optimising an objective function subject to constraints on the variance of estimated parametric functions. The concluding chapter explores future directions in the development of adaptive designs.

Randomised Response-Adaptive Designs in Clinical Trials presents methods for the randomised allocation of treatments to patients in sequential clinical trials. Emphasizing the practical application of clinical trial designs, the book is designed for medical and applied statisticians, clinicians, and statisticians in training. After introducing clin

This open access book presents robust statistical methods and procedures through the monitoring approach, with an emphasis on applications to linear regression. Illustrating the theory, it explores both large and small-sample properties. The performance of the forward search and of the monitoring of static robust estimators for regression data are illuminated through numerous data analyses using MATLAB and R.The book describes the results of many years’ work of the authors in the development of powerful methods of robust regression analysis. Robust methods are designed to analyse contaminated data. The well-established static robust methods estimate model features, such as parameter estimates, assuming the amount of contamination in the data is known. These methods are described in detail in Chapter 2 for estimation in a simple sample. The extension to regression is presented in Chapter 3, with an emphasis on S-estimation and related procedures as well as on least trimmed squares. The monitoring methods of Chapter 4, including the forward search, find the appropriate level of robustness for each data set and so avoid biased estimation from the inclusion of outliers and inefficiency due to the deletion of uncontaminated observations. This analysis is followed by examples which illustrate the use of the interactive graphical analyses associated with the authors’ FSDA toolbox. Numerical comparisons of the size and power of outlier tests appear in Chapter 5. Later chapters illustrate applications to response transformation in regression and to non-parametric regression. Extensions of the robust multiple regression model include Bayesian, heteroskedastic, time series and compositional regression, together with the clustering of regression models. Finally, several approaches to model selection are investigated and robust analyses of regression data are presented that illustrate the use of the techniques introduced earlier. Exercises are given at the end of each chapter, with solutions at the end of the book. The MATLAB code can be reproduced using MATLAB Online, without the need for a license, or via the language-agnostic Jupyter notebook environment, after installing the MATLAB kernel. Online computer code is available for all examples and exercises, together with a series of YouTube videos.Aimed at professional statisticians and researchers concerned with insightful data analysis, as well as postgraduate students, the book may also serve as a text for a modern interactive robust regression course.

This open access book presents robust statistical methods and procedures through the monitoring approach, with an emphasis on applications to linear regression. Illustrating the theory, it explores both large and small-sample properties. The performance of the forward search and of the monitoring of static robust estimators for regression data are illuminated through numerous data analyses using MATLAB and R.The book describes the results of many years’ work of the authors in the development of powerful methods of robust regression analysis. Robust methods are designed to analyse contaminated data. The well-established static robust methods estimate model features, such as parameter estimates, assuming the amount of contamination in the data is known. These methods are described in detail in Chapter 2 for estimation in a simple sample. The extension to regression is presented in Chapter 3, with an emphasis on S-estimation and related procedures as well as on least trimmed squares. The monitoring methods of Chapter 4, including the forward search, find the appropriate level of robustness for each data set and so avoid biased estimation from the inclusion of outliers and inefficiency due to the deletion of uncontaminated observations. This analysis is followed by examples which illustrate the use of the interactive graphical analyses associated with the authors’ FSDA toolbox. Numerical comparisons of the size and power of outlier tests appear in Chapter 5. Later chapters illustrate applications to response transformation in regression and to non-parametric regression. Extensions of the robust multiple regression model include Bayesian, heteroskedastic, time series and compositional regression, together with the clustering of regression models. Finally, several approaches to model selection are investigated and robust analyses of regression data are presented that illustrate the use of the techniques introduced earlier. Exercises are given at the end of each chapter, with solutions at the end of the book. The MATLAB code can be reproduced using MATLAB Online, without the need for a license, or via the language-agnostic Jupyter notebook environment, after installing the MATLAB kernel. Online computer code is available for all examples and exercises, together with a series of YouTube videos.Aimed at professional statisticians and researchers concerned with insightful data analysis, as well as postgraduate students, the book may also serve as a text for a modern interactive robust regression course.

This book collects the proceedings of the 10th Workshop on Model-Oriented Design and Analysis (mODa). A model-oriented view on the design of experiments, which is the unifying theme of all mODa meetings, assumes some knowledge of the form of the data-generating process and naturally leads to the so-called optimum experimental design. Its theory and practice have since become important in many scientific and technological fields, ranging from optimal designs for dynamic models in pharmacological research, to designs for industrial experimentation, to designs for simulation experiments in environmental risk management, to name but a few. The methodology has become even more important in recent years because of the increased speed of scientific developments, the complexity of the systems currently under investigation and the mounting pressure on businesses, industries and scientific researchers to reduce product and process development times. This increased competition requires ever increasing efficiency in experimentation, thus necessitating new statistical designs. This book presents a rich collection of carefully selected contributions ranging from statistical methodology to emerging applications. It primarily aims to provide an overview of recent advances and challenges in the field, especially in the context of new formulations, methods and state-of-the-art algorithms. The topics included in this volume will be of interest to all scientists and engineers and statisticians who conduct experiments.

This book collects the proceedings of the 10th Workshop on Model-Oriented Design and Analysis (mODa). A model-oriented view on the design of experiments, which is the unifying theme of all mODa meetings, assumes some knowledge of the form of the data-generating process and naturally leads to the so-called optimum experimental design. Its theory and practice have since become important in many scientific and technological fields, ranging from optimal designs for dynamic models in pharmacological research, to designs for industrial experimentation, to designs for simulation experiments in environmental risk management, to name but a few. The methodology has become even more important in recent years because of the increased speed of scientific developments, the complexity of the systems currently under investigation and the mounting pressure on businesses, industries and scientific researchers to reduce product and process development times. This increased competition requires ever increasing efficiency in experimentation, thus necessitating new statistical designs. This book presents a rich collection of carefully selected contributions ranging from statistical methodology to emerging applications. It primarily aims to provide an overview of recent advances and challenges in the field, especially in the context of new formulations, methods and state-of-the-art algorithms. The topics included in this volume will be of interest to all scientists and engineers and statisticians who conduct experiments.

This volume contains pioneering contributions to both the theory and practice of optimal experimental design. Topics include the optimality of designs in linear and nonlinear models, as well as designs for correlated observations and for sequential experimentation. There is an emphasis on applications to medicine, in particular, to the design of clinical trials. Scientists from Europe, the US, Asia, Australia and Africa contributed to this volume of papers from the 11th Workshop on Model Oriented Design and Analysis.

This book includes many of the papers presented at the 6th International workshop on Model Oriented Data Analysis held in June 2001. This series began in March 1987 with a meeting on the Wartburg near Eisenach (at that time in the GDR). The next four meetings were in 1990 (St Kyrik monastery, Bulgaria), 1992 (Petrodvorets, St Petersburg, Russia), 1995 (Spetses, Greece) and 1998 (Marseilles, France). Initially the main purpose of these workshops was to bring together leading scientists from ''Eastern'' and ''Western'' Europe for the exchange of ideas in theoretical and applied statistics, with special emphasis on experimental design. Now that the sep­ aration between East and West is much less rigid, this exchange has, in principle, become much easier. However, it is still important to provide opportunities for this interaction. MODA meetings are celebrated for their friendly atmosphere. Indeed, dis­ cussions between young and senior scientists at these meetings have resulted in several fruitful long-term collaborations. This intellectually stimulating atmosphere is achieved by limiting the number of participants to around eighty, by the choice of a location in which communal living is encour­ aged and, of course, through the careful scientific direction provided by the Programme Committee. It is a tradition of these meetings to provide low cost accommodation, low fees and financial support for the travel of young and Eastern participants. This is only possible through the help of sponsors and outside financial support was again important for the success of the meeting.

This volume contains the majority of the papers presented at the 5th Inter­ national Workshop on Model-Oriented Data Analysis held in June 1998. This series started in March 1987 with a meeting on the Wartburg, Eisenach (Germany). The next three meetings were in 1990 (St Kyrik monastery, Bulgaria), 1992 (Petrodvorets, StPetersburg, Russia) and 1995 (Spetses, Greece). The main purpose of these workshops was to bring together lead­ ing scientists from ''Eastern'' and ''Western'' Europe for the exchange of ideas in theoretical and applied statistics, with special emphasis on experimen­ tal design. Now that the separation between East and West has become less rigid, this dialogue has, in principle, become much easier. However, providing opportunities for this dialogue is as vital as ever. MODA meetings are known for their friendly atmosphere, leading to fruitful discussions and collaboration, especially between young and senior scien­ tists. Indeed, many long term collaborations were initiated during these events. This intellectually stimulating atmosphere is achieved by limiting the number of participants to around eighty, by the choice of location so that participants can live as a community, and, of course, through the care­ ful selection of scientific direction made by the Programme Committee.

The Fifth International Workshop on Model-Oriented Data Analysis (MODA5) focused on experimental design, particularly optimum design. A strength of this series of workshops is that they bring together leading scientists from "Eastern" and "Western" Europe. The proceedings therefore provides a valuable reference to the work of groups from many countries. In addition to 11 papers on optimum designs for linear and nonlinear models, there are groups of papers on designs for quality improvement, designs in agriculture and for model building. Non-design problems include robustness in linear models and estimation problems in nonlinear models. The volume concludes with a discussion on the teaching of experimental design.

This book includes many of the papers presented at the 6th International workshop on Model Oriented Data Analysis held in June 2001. This series began in March 1987 with a meeting on the Wartburg near Eisenach (at that time in the GDR). The next four meetings were in 1990 (St Kyrik monastery, Bulgaria), 1992 (Petrodvorets, St Petersburg, Russia), 1995 (Spetses, Greece) and 1998 (Marseilles, France). Initially the main purpose of these workshops was to bring together leading scientists from 'Eastern' and 'Western' Europe for the exchange of ideas in theoretical and applied statistics, with special emphasis on experimental design. Now that the sep­ aration between East and West is much less rigid, this exchange has, in principle, become much easier. However, it is still important to provide opportunities for this interaction. MODA meetings are celebrated for their friendly atmosphere. Indeed, dis­ cussions between young and senior scientists at these meetings have resulted in several fruitful long-term collaborations. This intellectually stimulating atmosphere is achieved by limiting the number of participants to around eighty, by the choice of a location in which communal living is encour­ aged and, of course, through the careful scientific direction provided by the Programme Committee. It is a tradition of these meetings to provide low cost accommodation, low fees and financial support for the travel of young and Eastern participants. This is only possible through the help of sponsors and outside financial support was again important for the success of the meeting.

The volume contains the proceedings of the 9th Workshop on Model-Oriented Design and Analysis. It offers leading and pioneering work on optimal experimental designs, both from a theoretical point of view and in respect of real applications. Scientists from Eastern and Western Europe, the US, Asia and Africa, are contributors to this volume. Designs for nonlinear models, for discriminating between models, for multivariate data and for applications to medicine are some of the primary topics covered.