Orador: Miguel de Carvalho (Pontificia Universidad Católica de Chile) Data, hora e local: 9 de dezembro de 2015, às 14h30m na sala de reuniões do Departamento de Matemática.
Resumo: Modeling nonstationarity in marginal distributions has been the focus of much recent literature in applied extreme value modeling. The simplest approach was popularized long ago by Davison and Smith (1990), and it is based on indexing the location and scale parameters of the generalized extreme value distribution by a predictor. And how to model ‘nonstationary multivariate/spatial extremes’ if one must? Surprisingly, by comparison to the marginal case, approaches to modelling nonstationarity in the extremal dependence structure have received relatively little attention.
In this talk, I will discuss approaches for modeling nonstationary extremal dependence structures. The approaches to be discussed can be regarded as natural extensions of the Davison–Smith paradigm to the multivariate and spatial contexts.
Biografia do Orador: Miguel de Carvalho is Assistant Professor of Applied Statistics, Pontificia Universidad Católica de Chile. Before moving to Chile he was a post-doctoral fellow at EPFL—Swiss Federal Institute of Technology. He is an applied mathematical statistician with a variety of interdisciplinary interests, inter alia, Biostatistics, Econometrics, and Statistics of Extremes. Beyond serving at the academy he is also a regular academic consultant of Banco de Portugal (Portuguese Central Bank). He his currently an Associate Editor of the Annals of Applied Statistics (IMS), and of Statistics and Public Policy (ASA).