Harry Joe, SSC Gold Medalist 2016
The Gold Medal is awarded to a person who has made outstanding contributions to statistics, or to probability, either to mathematical developments or in applied work.
The 2016 recipient of the Gold Medal of the Statistical Society of Canada is Professor Harry Joe.
Harry Joe is Professor in the Department of Statistics at the University of British Columbia. He received a BSc in Mathematics from the University of Victoria in 1978, an MSc in Statistics from the University of British Columbia in 1979, and a PhD in Statistics from Florida State University in 1982. His thesis, “Percentile Residual Life Function - Properties, Testing and Estimation,” was written under the supervision of Frank Proschan. He joined UBC as a faculty member in 1982, earning promotion to Full Professor in 1993. Over his career he has spent extended research visits at University College London, at the University of Pittsburgh and at the Technische Universität Muenchen where he held a Von Neumann Visiting Professorship.
Harry is internationally renowned for his work on multivariate dependence modeling. In addition to his journal articles in this area, Harry’s 1997 research monograph, Multivariate Models and Dependence Concepts, has become the `go-to’ reference for multivariate dependence. Some of the most influential ideas promoted here include inference functions for margins, analysis of tail dependence for multivariate non-Gaussian distributions and specific copula constructions. It was also the first book to bring together a wide range of dependence concepts in one place. The impact of this work is attested to by the more than 4400 citations (in Google Scholar) received to date. More recently, Harry has played a central role in the development of ‘vine copula’ models that have attracted considerable research attention. He is both the co-editor and the author of multiple chapters for the volume Dependence Modeling: Vine Copula Handbook, published in 2011. Then, in 2014, Harry published a monograph titled Dependence Modeling with Copulas that has comprehensive statistical coverage of copula modeling. This promises to be as, or more, impactful than his earlier monograph. There is a companion software (large R package) that has algorithms and numerical methods for implementation of high-dimensional copula models.
Harry’s body of research is also extremely broad. Beyond his multivariate dependence work, he has published in areas such as statistical computing, paired-comparison models, time-series models and composite likelihood techniques. He has also worked across disciplinary boundaries. He has published several papers with psychometricians on limited information goodness-of-fit procedures for models for high-dimensional ordinal response, and multiple papers with geneticists that develop and apply statistical methods in the study of the genetic diseases called type 1 and type 2 neurofibromatosis.
At UBC Harry has supervised many graduate students and visiting students and has served the department and university in numerous ways, including a term as Department Head. Beyond UBC Harry has served the SSC through several committee memberships and has served the research community via many editorial roles.
The citation for the award reads:
"To Harry Joe, in recognition of his deep and impactful contributions to statistical science, most notably in multivariate dependence modelling, but also in composite likelihood, statistical computing, time-series analysis, and methodology for psychometric and genetic applications."