Robert Tibshirani, SSC Gold Medalist 2012

Robert Tibshirani
SSC Gold Medalist

Robert John Tibshirani, a professor in the Department of Health Research and Policy and in the Department of Statistics at Stanford University is the winner of the 2012 Gold Medal of the Statistical Society of Canada. This award is the highest distinction bestowed by the SSC. It is given annually to a Canadian statistician or probabilist who has made outstanding research contributions to statistical sciences and is intended to honor a leader in the field.

Rob Tibshirani studied statistics and computer science at the University of Waterloo (BMath, 1979), University of Toronto (MSc, 1980) and Stanford University (PhD, 1984). His PhD thesis was written under the supervision of Bradley Efron. He was an Assistant Professor (1985-89), Associate Professor (1989-94) and Professor (1994-98) at the University of Toronto. In 1998 he joined Stanford University as Professor.

Rob's contributions in the statistical sciences place him in the uppermost echelon of researchers worldwide. He has made exceptional contributions to methodology and theory for the analysis of complex data sets, smoothing and regression methodology, statistical learning, and classification, and application areas that include public health, genomics, and proteomics. Some of his best-known contributions include the Lasso, which uses absolute value penalization in regression and related problems, Generalized Additive Modeling, and Significance Analysis of Microarrays (SAM). He has co-authored three widely used books Generalized Additive Models, An Introduction to the Bootstrap, and The Elements of Statistical Learning, now in its second edition. These books are very widely used not only in statistics but also in other fields and their material is taught in graduate schools around the world.

Rob has published over 240 refereed papers in leading statistics journals such as The Annals of Statistics, Biometrika, Biometrics, Biostatistics, The Canadian Journal of Statistics, Journal of the American Statistical Association, and Journal of the Royal Statistical Society, Series B. He has also published widely in other leading scientific journals including Bioinformatics, New England Journal of Medicine, Science, Neural Computation, and Breast Cancer Research and Treatment.

Rob's earliest work was on the bootstrap, where he wrote major papers with Bradley Efron on confidence intervals, estimation of prediction error, and model search. He soon moved into other computer- intensive methods for estimation, clustering, and classification with a number of landmark papers such as his 1986 Statistical Science and 1987 JASA papers on generalized additive models with Trevor Hastie. In 1996 he published a paper in JRSS-B in which the Lasso was born. Many of his major contributions overlap the field of machine learning in computer science, and Rob, Jerry Friedman, and Trevor Hastie began to use the term statistical learning, which is in the title of their 2001 book.

Rob's other contributions are too numerous to list, but major areas include adaptive logistic regression and boosting (with Friedman and Hastie, The Annals of Statistics, 2000), and least angle regression (with Efron, Hastie, and Johnstone, The Annals of Statistics, 2004).

Rob has also made exceptionally broad and important contributions to genetics, medicine, public health, traffic safety, and other scientific areas. His current research focuses on the analysis of high-dimensional data, with a special focus on applications in genomics, proteomics, and computational biology. Rob co-authored the first study linking cell phone usage with car accidents; this widely cited article played a role in the introduction of legislation that restricts the use of phones while driving. He is one of the most widely cited authors in the entire mathematical sciences. On the lighter side, in his paper "Who is the fastest man in the world?" he analyzed sprint races after the 1996 Olympics and correctly predicted (Canadian) Donovan Bailey's victory over Michael Johnson in the 1997 special match race at Toronto's SkyDome.

Rob's exceptional contributions in research have been recognized through various awards and honors. He is a Fellow of the Royal Society of Canada, the American Statistical Association, and the Institute of Mathematical Statistics. He won the worldwide COPSS Award in 1996, the NSERC Steacie Award in 1997, the CRM-SSC Prize in Statistics in 2000, the University of Waterloo Distinguished Alumni Achievement Award in 2006, and he was an IMS medallion lecturer.

Rob has also served the profession in many capacities: as Associate Editor for many journals including The Canadian Journal of Statistics, as program chair for conferences such as the SSC meeting, and through memberships in professional committees and panels for organizations such as ASA, IMS and NSERC. He has been heavily involved in PhD supervision, graduating 29 PhD students who have gone on to significant careers in academia and industry.

His parents, Sami and Vera Tibshirani, have given him a lifetime of love and support. He has a wonderful wife Cheryl and they are blessed with three special children: Charlie, Ryan, and Julie. Ryan is also one of his proudest "contributions" to the field of statistics: he is now an Assistant Professor in Statistics at Carnegie Mellon University.

Rob received his Gold Medal at the 40th SSC Annual meeting held in Guelph, Ontario, June 3 to 6, 2012.

The citation for the award reads: 

"To Robert John Tibshirani, for pioneering work in the development and implementation of statistical methodology in many important and evolving fields such as the bootstrap, generalized additive models, statistical learning, high-dimensional data analysis, multiple hypothesis testing, significance analysis of microarrays, and for broad and important contributions to genetics, medicine, public health, traffic safety and other scientific areas."

Rob will deliver the Gold Medal Address at the 41st Annual Meeting of the Society to be held May 26 to 29, 2013 in Edmonton, Alberta.