[ABE-L] Ciclo de Palestras=?ISO-8859-1?Q?_do_PPG_em_Estat=EDs?=tica (=?ISO-8859-1?Q?15_de_mar=E7o?=)
Fernando A. S. Moura
fmoura em im.ufrj.br
Qui Mar 9 13:19:55 -03 2017
Caros,
Nesta quarta-feira, dia 15 de março, excepcionalmente às 13:30hs no
Laboratório de Sistemas Estocásticos (LSE), sala I-044b, o Professor
Havard Hue (KAUST) dará uma palestra pelo Ciclo de Palestras do PPG em
Estatística.
Os detalhes seguem abaixo.
Mais informações sobre o Ciclo de Palestras do PPG em Estatística
podem ser encontradas no site www2.dme.ufrj.br.
Ciclo de Palestras do PPG em Estatística - IM-UFRJ
Data: 15 de março de 2017
Hora: 13:30
Local: LSE
Palestrante: Havard Hue (KAUST)
Título: Penalising model component complexity: A principled practical
approach to constructing priors
Setting prior distributions on model parameters is the act of
characterising the nature of our uncertainty and has proven a critical
issue in applied Bayesian statistics. Although the prior distribution
should ideally encode the users uncertainty about the parameters, this
level of knowledge transfer seems to be unattainable in practice and
applied statisticians are forced to search for a default prior. Despite
the development of objective priors, which are only available explicitly
for a small number of highly restricted model classes, the applied
statistician has few practical guidelines to follow when choosing the
priors. An easy way out of this dilemma is to re-use prior choices of
others, with an appropriate reference.
In this talk, I will introduce a new concept for constructing prior
distributions. We exploit the natural nested structure inherent to many
model components, which defines the model component to be a flexible
extension of a base model. Proper priors are defined to penalise the
complexity induced by deviating from the simpler base model and are
formulated after the input of a user-defined scaling parameter for that
model component, both in the univariate and the multivariate case. These
priors are invariant to reparameterisations, have a natural connection to
Jeffreys priors, are designed to support Occams razor and seem to have
excellent robustness properties, all which are highly desirable and allow
us to use this approach to define default prior distributions. Through
examples and theoretical results, we demonstrate the appropriateness of
this approach and how it can be applied in various situations, like random
effect models, spline smoothing, disease mapping, Cox proportional hazard
models with time-varying frailty, spatial Gaussian fields and multivariate
probit models, etc. Further, we show how to control the overall variance
arising from many model components in hierarchical models.
This is joint work with a lot of people related to the R-INLA project, and
is still work in progress.
Atenciosamente
Fernando Moura
Mais detalhes sobre a lista de discussão abe