Year: 2025
Type: article
Abstract: David Draper died at the age of 72, on 8 May 2024, in Santa Cruz, California. David was a distinguished and pioneering statistician, an amazing teacher and an inspiring supervisor, whose contributions to the field have left an indelible mark. He was a true expert in analysing complex data and gained great satisfaction in successfully addressing real-life challenges in a variety of fields. In 2002, he served as President of the International Society for Bayesian Analysis (ISBA), and since 2014, he was made an ISBA fellow for his fundamental Bayesian research on model uncertainty, hierarchical modelling, and applications of Bayesian decision theory in complicated applied settings. In addition, he made significant contributions to applied Bayesian problem-solving in education, environmental risk assessment, health policy and medicine, and gave significant service to ISBA and the Bayesian community more widely including extensive Bayesian educational activities. He was born in Omaha, Nebraska and raised in Temple Hills, Maryland. David’s interest in statistics began from an early age. He studied Mathematics at the University of North Carolina, Chapel Hill, from 1970 to 1974 and was awarded with highest honours. After working for IBM for a few years, in 1980 he received an MA in Statistics, from the University of Berkeley in California and in 1981 he completed a PhD in Statistics, from the same institution, under the supervision of Erich Lehmann. After his PhD, David began his academic career when in 1981 he joined the Department of Statistics in the University of Chicago as a Research Associate. In 1984, he moved to the RAND Corporation, in Santa Monica, California to the Statistical Research and Consulting Group, initially under the Department of Economics and Statistics and thereafter under the Graduate School of Policy Studies. From 1990, he joined the Division of Statistics in UCLA and in 1992 he took on the position of Associate Director of the Statistical Consulting Centre in the same institution. In 1993, he moved to the UK where he was appointed Lecturer in Statistics, within the Statistics Group of the Department of Mathematical Sciences at the University of Bath. At Bath, he was promoted to a Readership in 1995 and awarded his first chair in 1999. After 8 years in the UK, he moved back to the U.S. in 2001, to the University of California at Santa Cruz (UCSC), as the founding Chair at the Department of Applied Mathematics and Statistics in the Baskin School of Engineering where was also Professor of Statistics. In addition, he taught and worked on research projects at the University of Chicago, the University of Washington, and the University of Neuchatel (Switzerland). He was a fellow of the American Association for the Advancement of Science, the American Statistical Association, the Institute of Mathematical Statistics, and the Royal Statistical Society. David’s research work was both prolific and influential. His research concerned methodological developments in Bayesian statistics, with particular emphasis on hierarchical modelling, Bayesian nonparametric methods, model specification and model uncertainty, quality assessment, risk assessment and applications in the environmental, medical, and social sciences. His outstanding 1995 paper on Assessment and Propagation of Model Uncertainty, which was published as a discussion article in the Journal of the Royal Statistical Society, Series B, has received more than 2400 citations from researchers working in many different disciplines. In this methodological article, he documents, in an exceptionally clearly written way, the failure of standard empirical model-building methods to capture the uncertainty in the modelling process itself and demonstrates the success of a Bayesian approach to solving the problem. After 28 years, this paper is still a joy to read and continues to inspire many researchers in various fields. He was the author or co-author of more than 100 contributions to the research literature, including books, journal articles, letters, book chapters, book reviews, encyclopaedia articles, invited discussions, and contributed discussions, and during his academic career, he received more than $5 million in grants to fund his research as a PI or co-PI. He was an Associate Editor for six leading journals, and he organized or co-organized six international research conferences. He gave a significant number of invited, special invited and plenary talks at major research conferences and leading statistics departments, and his short courses were highly regarded. David was a charismatic, inspiring, dedicated, and skilled teacher and supervisor. His teaching philosophy was to excite students about the topic, using real-life examples and applications and his teaching was always extremely interactive, since he strongly believed that students should be an active part of the teaching process. His energy and enthusiasm were infectious, he had a unique way to make the complex appear simple and was always willing to spend more time than was scheduled to explain ideas and discuss them with students; in return, he was loved by many students who regularly in student feedback described David as the best professor they were ever taught by. During his academic career, he received numerous awards for Excellence in Teaching, from the University of Chicago, from the RAND Graduate School of Policy Studies, from the University of Bath and from the University of California at Santa Cruz (the last one only 2 years ago!) and fortunately David never stopped teaching. In 1998 and 2004, he received an Excellence in Continuing Education award from the American Statistical Association. He mentored nine graduate students to PhD degrees at universities in Sweden, the UK and the U.S., and he supervised three post-doctoral research associates at Bath and UCSC. David had a strong passion for Bayesian Statistics. During his academic career, he gave more than 40 short courses on Bayesian methods and applications in Brazil, Canada, Finland, Greece, India, New Zealand, Switzerland, the U.K., and the U.S. He used to call himself a de Finetti-style Bayesian, since he was arguing that prediction of observables is more fundamental than inference about unobservables and (conditional) exchangeability judgments are more fundamental to predictive modelling. Beyond his remarkable academic career, he was a man of many passions. He loved travelling, always eager to explore new places and cultures. He deeply appreciated a good meal, whether he was cooking it himself or enjoying it in good company in a local pub or a five-star restaurant. He was a thoughtful and engaged citizen of the world, he followed politics closely and believed profoundly in human freedom, equality, and justice. He stood by minorities and those in need, always ready to offer help, guidance, or simply a compassionate ear. He was a great listener—patient, empathetic, and relentlessly committed to finding solutions, no matter how complex the problem was. He was friendly and generous by nature, he found joy in watching sports, sharing meals, and engaging in lively conversations with the people he cared about. His warmth, curiosity, and unwavering sense of humanity touched everyone who had the privilege to know him. His memory and significant contributions will continue to inspire and guide future generations of statisticians. For me David was my academic father, he was the person that persuaded me to follow the path I am currently on, and I owe him much of my success. He taught me in a vigorous way, how to be a good teacher, how to do research and how to analyse data. I will deeply miss working together with David, our interesting discussions and, above all, I will miss our friendship. David, you will be profoundly missed! Thank you for giving us all so much! No new data were generated or analysed in support of this research.
Author: Dimitris Fouskakis
Institution: National Technical University of Athens
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4413106173