[Prévia] [Próxima] [Prévia por assunto] [Próxima por assunto]
[Índice cronológico]
[Índice de assunto]
Conferências
- Subject: Conferências
- From: Marcia Branco <mbranco@ime.usp.br>
- Date: Thu, 27 Nov 2008 10:29:23 -0200
Colegas
O programa de pós-graduação em estatística do IME-USP tem a satisfação
de convidá-lo
para as conferências dos professores Hedibert Lopes (15 horas) e Michael
Höhle (16 horas)
na próxima dia 01/12 (segunda-feira), sala 136 do bloco A do IME.
Atenciosamente,
Fábio Machado e
Márcia Branco
Seguem os resumos.
----------------------------------------------------------------------------------------------------------------------
Dr. Hedibert Freitas Lopes
University of Chicago
Title: Particle Learning and Smoothing
ABSTRACT: This paper provides novel particle learning (PL) methods for
sequential filtering, parameter learning and
smoothing in a general class of state space models. The approach extends
existing particle methods by incorporating
unknown fixed parameters, utilizing sufficient statistics, for the
parameters and/or the states, and allowing for nonlinearities
in the model. We also show how to solve the state smoothing problem,
integrating out parameter uncertainty. We show that
our algorithms outperform MCMC, as well as existing particle filtering
algorithms.
This paper is joint work with C.Carvalho, M.Johannes and N.Polson.
-------------------------------------------------------------------------------------------------------------------------------
Dr Michael Höhle
Department of Statistics, Ludwig-Maximilians-
University Munique, Alemanha.
Title: Stochastic Spatio-Temporal Epidemic Modelling
ABSTRACT: Mathematical modelling is an important tool in order to
better understand the dynamics of infectious
diseases - be it in human or animals. A key epidemic model is here the
stochastic susceptible-exposed-infectious (SIR) model.
This presentation discusses work on extending the SIR modelling with
additional focus on the spatial dimension by attempting a regression
view on disease dynamics. A continuous time stochastic process model
based on conditional intensities is proposed, where dynamics is split
into epidemic and endemic components. Simulation from the model using
Ogata's modified thinning algorithm is discussed together with likelihood
based model inference and residual analysis.
As illustration, data provided by the Federal Research Centre for Virus
Diseases of Animals, Wusterhausen, Germany, on the incidence of
classical swine fever virus in Germany during 1993-2004 are analysed.