British statistician John Brian Copas will be awarded the Akaike Memorial Lecture Award, the Japanese statistical research community's major prize, for his many contributions to the study of data bias in the fields of medical statistics, econometrics and psychometrics. He is honored in particular for his work on the risk of re-conviction in the field of criminal justice, and for shedding light on the issue of publication bias in the estimation of cancer risk from passive smoking.
John Brian Copas, Professor Emeritus at the University of Warwick, UK will receive the Akaike Memorial Lecture Award from the Institute of Statistical Mathematics (ISM) and the Japan Statistical Society (JSS) in recognition of his considerable and wide-ranging achievements in statistical methodology with a focus on practical applications, especially in regard to various sources of data bias within medical statistics, econometrics and psychometrics, for his work on risk of re-conviction with the field of criminal justice, and for his research into publication bias in the estimation of cancer risk from passive smoking. This is the third time the award has been given.
Professor Copas presented his doctoral thesis in 1967 on the compound decision problem to the Imperial College London under the supervision of Professor George Alfred Barnard. In 1969, he published a paper with discussion on compound decisions and empirical Bayes a in the Journal of the Royal Statistical Society (JRSS) . This paper provided a penetrating insight into decision-making problems in the case where each observation has an unknown parameter. This work laid the foundation of his famous 1983 article, "Regression, Prediction and Shrinkage" . The 1983 paper shed light on Stein's seemingly strange phenomenon in estimating means (Stein's paradox) from the viewpoint of prediction, expanding statistical understanding of the shrinkage behavior. Excellent works on shrinkage estimation theory have been conducted by many Japanese statisticians, and Professor Copas' 1983 article is considered to have served as a springboard for their achievements. Professor Copas's six papers were read before the Royal Statistical Society and published with discussions in JRSS. His works are always achieved on a fine balance between theory and application. Moreover, his research efforts provided a deep insight and wide perspective on various sources of data bias , making a lasting impact for years to come. His research interests are wide-ranging, and his contributions have spanned academic disciplines such as medical statistics, econometrics and psychometrics.
For example, he co-authored an article that revisited meta-analytical evidence on lung cancer and passive smoking . This work showed possible involvement of publication bias in the estimation of cancer risk, and suggested that the epidemiology community should reconsider the conventional approaches to meta-analysis and interpretation. The Copas selection model, developed for sensitivity analysis in this work, has been a common method for assessing the impact of publication bias in meta-analysis .
Moreover, Professor Copas' long-term achievements include criminal statistics analysis. With the rising sex, drug, and other offence rates resulting from rapid social changes in the UK, he made a series of contributions, as a statistician, to the understanding and prediction of delinquency. In particular, his approaches using Cox model analysis for re-offending events and logistic regression analysis for re-conviction and risk factors were highly appreciated by the Ministry of Justice experts, and the "Copas rate," a re-offence probability estimator, has been in reports by the Ministry of Justice .
His research stance reflects the rich tradition of the British school of statistics. He has extended open arms and accepted students and researchers from Japan, China, Iran, and other countries.
Since Professor Copas retired from the University of Warwick, he has continued his research activities in affiliation with the University College London and other institutions. In those years, he has been vigorously engaged in multivariate meta-analysis, network meta-analysis, and other topics of practical significance together with many British front-line scientists. His unrelenting enthusiasm is an inspiration for many researchers.
 Copas, J. B. (1969).
Compound Decisions and Empirical Bayes (with discussion).
Journal of the Royal Statistical Society, B, 31, 397-425.
 Copas, J. B. (1983).
Regression, Prediction and Shrinkage (with discussion).
Journal of the Royal Statistical Society, B, 45(3), 311-354.
 Copas, J.B. and Li, H.G. (1997).
Inference for non-random samples (with discussion).
Journal of the Royal Statistical Society, B, 59, 55-95.
 Copas, J.B. and Eguchi, S. (2005).
Local model uncertainty and incomplete-data bias (with discussion).
Journal of the Royal Statistical Society, B, 67, 459-513.
 Copas, J. B. and Shi, J.Q. (2000).
Reanalysis of Epidemiological Evidence on Lung Cancer and Passive Smoking.
British Medical Journal, 320 (7232) 417-418.
 Copas, J.B. and Shi, J.Q. (2000).
Meta-analysis, funnel plots and sensitivity analysis.
Biostatistics, 1, 247-262.
 Copas, J.B. and Marshall, P. (1998).
The Offender Group Reconviction Scale: A Statistical Reconviction Score for Use by Probation Officers.
Journal of the Royal Statistical Society, C, 47, 159-171.
Reasons for Award
In the 1980s, he developed a novel concept for expanding the Akaike Information Criterion model selection method (whose developer gives the Akaike Award its name), and proposed a method to consider measured and predictive distributions separately.
Professor Copas's contributions in the area of statistical inference are particularly salutary. The Rubin causal model, proposed in the 1970s by Professor Donald Rubin at Harvard University, is one of the most well-known statistical frameworks for causal inference. But it was Professor Copas's efforts regarding statistical causal inference that laid the foundation for Rubin with respect to how epidemiologic research prompted mathematical arguments about proliferation of attributable risk (Hamilton's model).
The research efforts of Professor Copas have always achieved a fine balance between theory and application. Above all, his research efforts have provided deep insights into data bias in a wide range of areas, from medical statistics to econometrics and psychometrics.
Biography of Professor John Brian Copas
John Brian Copas, born in 1943, is Professor Emeritus at the University of Warwick. With a focus on both theory and application, his works provide a deep insight and wide perspective on various sources of data bias. His contributions span over a wide range of academic disciplines, including statistical medicine, ecometrics, and pscyhometrics. He has developed the Copas selection model, a well-known method for sensitivity analysis, and the Copas rate, a re-conviction risk estimator used in the criminal justice system.
1967 Imperial College London: PhD Statistics
1966-1973 Lecturer/Senior Lecturer, University of Essex, UK
1969-1970 Assistant Professor, State University of New York at Buffalo, USA
1973-1983 Professor of Statistics, University of Salford, UK
1983-1991 Professor of Statistics, University of Birmingham, UK
1992- Professor of Statistics, University of Warwick, UK
2009 - Emeritus Professor at the University of Warwick, UK
2019- Honorary Professor at University College London, UK
Served two three-year terms as head of the Department of Statistics at Warwick.
Also short-term visiting appointments at several overseas universities and research institutes (including the ISM).
Overview of the Akaike Memorial Lecture Award
The Akaike Memorial Lecture Award has been planned since 2014 under the joint sponsorship of ISM and JSS. We have named this lecture award after Dr. Hirotugu Akaike, who left a wide-reaching and influential legacy of research in the statistical sciences, and intend for these events to be both opportunities for exchange among statistical researchers from inside and outside Japan and to provide inspiration to young and talented researchers, contributing to further advances in this field.
Every two years, one lecturer is selected under the standards of the Akaike Memorial Lecture Award Nominating Committee from among those individuals who have, like Dr. Akaike, stood out as being ahead of their time, exercising an international influence over a wide range of fields in the statistical sciences (including mathematical sciences such as control and optimization, and mathematical engineering) and applied fields. The awardee receives an honorarium (100,000 yen), an award plaque, and travel expenses.
To promote the education of students and young researchers, the Akaike Memorial Lecture features a selected board of representatives who will engage in discussions after the lecture. The lecture and follow-up discussion will be published as an invited paper in the Annals of the Institute of Statistical Mathematics (AISM) or the Journal of the Japan Statistical Society (JJSS).
The 2020 Akaike Memorial Lecture
Speaker: Prof. John Brian Copas
(University of Warwick, Department of Statistics)
Title: Some of the Challenges of Statistical Applications
Discussants: Professor Masataka Taguri (Yokohama City University) and Dr. Masayuki Henimi (The Institute of Statistical Mathematics)
Date and Time: September 9, 2020, 16:30-18:00
Venue: Toyama International Conference Center
For more information, please visit: http://www.
About the Research Organization of Information and Systems (ROIS)
The Research Organization of Information and Systems (ROIS) is a parent organization of four national institutes (National Institute of Polar Research, National Institute of Informatics, the Institute of Statistical Mathematics and National Institute of Genetics) and the Joint Support-Center for Data Science Research. It is ROIS's mission to promote integrated, cutting-edge research that goes beyond the barriers of these institutions, in addition to facilitating their research activities, as members of inter-university research institutes.
About The Institute of Statistical Mathematics (ISM)
The Institute of Statistical Mathematics (ISM) is one of Japan's national research institutes and also a part of ROIS. ISM is an internationally renowned institute with more than 75 years of history, specialized in research on statistical mathematics. ISM pursues research and high quality education by its biaxial structure for research and education programs, three basic research departments, four NOE (Network Of Excellence)-type research centers, and the school for professional development. Through the efforts of its research departments and centers, ISM aims to continuously facilitate cutting edge research collaboration with universities, research institutions, and industries both from Japan and abroad.