SEMINÁRIO: Using Latent Gaussian Models to Infer the Spatial Structure Inherent in Point Patterns


  • Charlotte Jones-Todd – Centre for Research into Ecological and Environmental Modelling – University of St. Andrews – Scotland
  • FCUL – Campo Grande – Bloco C6 Piso 4 – Sala: 6.4.31 – 11:30h
  • Quarta-feira, 1 de Abril de 2015

Interest in understanding and inferring the underlying spatial mechanics of patterns formed by the locations of objects in space is not a recent concept, however the computational cost of fitting such complex models has historically limited their application to comparably simple models. Through the use of Integrated Nested Laplace Approximation (INLA) complex spatio-temporal models can be fitted to infer the inherent latent processes. By defining the underlying spatial field through a stochastic partial differential equation we exploit the concepts of a continuously defined random field; approximating it by a Gaussian Markov random field the procedure is implemented within the INLA model fitting frame-work.