Deming and Fisher Lecturers Announced
As the ASA membership gears up for JSM 2010, those planning to attend the annual meetings must decide which sessions to attend and which activities to undertake. There are several to choose from, but the keynote addresses are special. Each speaker was chosen specifically for his or her vast knowledge of statistics and dedicated work in the field. Here we introduce these speakers and hope you will consider attending their addresses.
Brent C. James, MD, M.Stat.
“Better: Dr. Deming Consults on Quality for Sir William Osler”
Tuesday, August 3, 4 p.m.
Brent James is the chief quality officer and executive director of the Institute for Health Care Delivery Research at Intermountain Healthcare in Salt Lake City. He is known internationally for his work in clinical quality improvement, patient safety, and the infrastructure that underlies successful improvement efforts, such as culture change, data systems, payment methods, and management roles.
Through the Intermountain Advanced Training Program in Clinical Practice Improvement (ATP), he has trained more than 3,500 senior physician, nursing, and administrative executives, drawn from around the world, in clinical management methods, with proven improvement results (and more than 30 “daughter” training programs in six countries).
Before coming to Intermountain, James was an assistant professor in the department of biostatistics at the Harvard School of Public Health, providing statistical support for the Eastern Cooperative Oncology Group (ECOG); and he staffed the American College of Surgeons’ Commission on Cancer.
He earned BS degrees in computer science (electrical engineering) and medical biology; an MD degree (with residency training in general surgery and oncology from the University of Utah); and a master of statistics degree.
A participating member of the National Academy of Science’s Institute of Medicine, James also is a fellow of the American College of Physician Executives and has won numerous awards including the Utah Business Magazine “Health Care Heroes” Lifetime Achievement Award in 2009.
He serves on several nonprofit boards of trustees dedicated to clinical improvement.
COPSS Fisher Lecture
“Likelihoods with Hidden Variables”
Wednesday, August 4, 4 p.m.
Hal S. Stern, Chair, COPSS Fisher Lecture Committee
The 2010 Council of Presidents of Statistical Societies (COPSS) has named Bruce Lindsay, Willaman Professor of Statistics at Pennsylvania State University, to deliver the Fisher Lecture at the Joint Statistical Meetings in Vancouver. Lindsay is a deserving awardee; as one of his supporters wrote, “Like Fisher himself, Bruce brings keen geometric insight to bear on problems of practical relevance.’’ He is well known for developing statistical theory that serves as a basis for sound statistical practice.
Lindsay was born in Oregon and earned his bachelor’s degree in mathematics at the University of Oregon in 1969. Following a stint in the U.S. Coast Guard, Lindsay completed his PhD in biomathematics at the University of Washington in 1978. Following a one-year postdoctoral appointment at Imperial College in London, he accepted an appointment in the department of statistics at Penn State. He has been there ever since; he was named a distinguished professor in 1991 and was named the Willaman Professor of Statistics in 2004.
Lindsay is a Fellow of the ASA and the Institute of Mathematical Statistics. He was previously recognized with the COPSS Snedecor Award in 1995–1996 for contributions to biometrics. He has more than 75 refereed publications in top statistics journals and has received continuous external support for his research from the National Science Foundation since 1980. He has served as the doctoral advisor for 26 PhD students, many of whom now serve as faculty members. Also, he has given distinguished lectures at numerous universities and conferences around the world.
There are numerous research contributions for which Lindsay is being recognized, almost all having to do with his body of work on likelihood-based inference. He has made a number of significant contributions to the study of mixture models. Mixture models are often used to represent a population as being comprised of a mixture of subpopulations. Lindsay has created a general geometric theory for the study of mixture models and a basic theory for nonparametric inference for such models. Mixture models are widely used in the social sciences and increasingly in bioinformatics and computational biology. Thus there is, in the Fisher tradition, a strong link to practical problems.
Lindsay also developed a key theory for addressing nuisance parameter bias, which has been used extensively in measurement error models. Measurement error models attempt to address the situation where we are unable to precisely measure a key variable, like food intake, that may be related to an outcome of interest, perhaps disease incidence. It is a challenging yet important problem to assess the impact of the variable while taking into account the variability associated with our measurement process.
A third significant contribution is Lindsay’s development of the composite likelihood. Composite likelihoods are used in problems for which a complete joint distribution for all of the variables of interest cannot be obtained. The composite likelihood is a form of approximation, and Lindsay’s theory helped develop some of the properties of the approximation. This work has had a significant impact on the development of spatial models, which are now prevalent in many domains of science.
Lindsay’s varied contributions to statistical theory underlie important practical methods used for spatial models, measurement error models, and mixture models across a wide range of scientific domains. These contributions mark Lindsay as an innovator in the Fisher style and make him a worthy choice to deliver the lecture named in Fisher’s honor.