Non-Gaussian Multivariate Statistical Models and their Applications (13w5146)

Arriving in Banff, Alberta Sunday, May 19 and departing Friday May 24, 2013


(McMaster University)

Chris Field (Dalhousie University)

(King Abdullah University of Science and Technology)

(University of British Columbia)


The objectives of this workshop are:

(1) to bring together researchers to address emerging issues in non-Gaussian
multivariate statistical models and their application;

(2) to promote interactions and collaborations among researchers in Canada and around the world on two specific lines of research, namely multivariate skew-symmetric distributions and copula models;

(3) to explore the combination of multivariate skew-symmetric distributions and copulas to obtain tractable models with tail dependence and reflection tail asymmetry;

(4) to discuss ideas and future directions for developing realistic and useful non-Gaussian statistical models, with emphasis on their applications;

(5) to provide opportunities for graduate students and recent Ph.D. recipients to meet and interact with leading researchers in the area.

This is the first-of-its-kind workshop that will develop interactions between the two communities of multivariate skew-symmetric distributions and copula models, and especially among young researchers. Each of those two distinct fields of approaches to multivariate non-Gaussian statistical modeling have now matured enough separately, so that it is very timely to organize such a workshop. It will create new interdisciplinary interactions and provide a forum for vigorous research oriented discussions. The field of non-Gaussian multivariate statistical models is very important in statistics and relevant because of its many applications, including areas such as economics, finance, management science, oceanography, climatology, environmetrics, forestry, engineering, renewable energies, image processing, astronomy, clinical trials, biomedical science, and many more. Moreover, several of the aforementioned applications are related to processes affecting Planet Earth. Therefore, they are also relevant to the Mathematics of Planet Earth - 2013 (MPE 13) initiative. We expect that this workshop will serve as a platform for significant further developments.

Each day of the workshop will include presentations of recent research in the theory and applications of non-Gaussian multivariate statistical models according to the following tentative schedule:

Monday: Skew-symmetric models: definitions, properties and inference.
Tuesday: Copula models: definitions, properties and inference.
Wednesday: Other non-Gaussian models.
Thursday: Applications of all those models (including to MPE 13) and software.
Friday: challenges and open problems.

The presentations will be followed by discussion of the most promising avenues for new research. We plan to have a discussant for each session and allow ample time for general discussions and interactions.