Collaborative Research Team Project: 2014-2017
Copulas have emerged as a tool for capturing and modeling dependence since the mid-1980s and have played an important role in applications ranging from finance to genetics to hydrology. The aim of the Collaborative Research Team is to provide new methodological developments including semi- and non-parametric strategies that will enable flexible inference of conditional dependencies in multivariate copula models. Existing data-driven methods for conditional copula models are suitable only in low-dimensional problems, and easily become impractical as dimension increases. To tackle computational challenges in high dimensions and to make inference feasible, the team will investigate dimension reduction techniques and variable-merging methods, as well as tests to identify conditional independence among groups of variables. The interpretation of dependence features in high-dimensional data is another challenge that requires a careful study of numerical and graphical representations.
The research has the potential to impact applications in finance and genetics in Canada and worldwide. To this end, the collaborators include world-class statistical scientists at universities and from government. As well, the team has forged important collaborations with industry partners including Banque Nationale du Canada, Bank of Montreal and Institut de recherche d’Hydro-Québec.
The team leaders are Louis-Paul Rivest of Université Laval and Christian Genest of McGill University. Their collaborators are Elif Acar of the University of Manitoba, Radu Craiu of the University of Toronto, Harry Joe of the University of British Columbia, Johanna Nešlehová of McGill University, Jean-François Quessy of UQTR, Bruno Rémillard of HEC Montréal, Claudia Czado of Technische Universität München, Anne-Catherine Favre of Laboratoire d’étude des transferts en hydrologie et environnement, Grenoble, Anne-Laure Fougères of Université de Lyon 1, and Marius Hofert of ETH Zürich.