<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <title></title> </head> <body style="font-family:Arial;font-size:14px"> Ciclo de Palestras do Programa de Pós-graduação em Estatística da UnB Divulgação de palestra do Programa de Pós-graduação em Estatística da UnB. Palestrante: Thais Cristina Oliveira da Fonseca (IM-UFRJ) Título: ?Dynamic Multiscale Spatiotemporal Models for Poisson Data.? DATA: 17/09/2015 (quinta-feira) HORÁRIO: 14:30h LOCAL: Sala Multiuso EST Resumo: We propose a new class of dynamic multiscale models for Poisson spatiotemporal data. The spatiotemporal multiscale coefficients evolve through time accordint to a novel Dirichlet evolution. We propose a simulation-based full Bayesian posterior analysis based on the posterior conditional independence of the multiscale coefficients. Thus, the proposed method is computationally efficient, and highly par- allelizable. We present analysis of discount factors as a tool to identify regions with distinctive spatiotemporal dynamics. Finally, we illustrate our multiscale spatiotemporal Poisson methodology with an application to tornado reports in the American Midwest. </body> </html>