A sequential Monte Carlo approach to the sequential design for discriminating between rival continuous data models


Autoria(s): McGree, James; Drovandi, Christopher C.; Pettitt, Anthony N.
Data(s)

21/09/2012

Resumo

Here we present a sequential Monte Carlo approach to Bayesian sequential design for the incorporation of model uncertainty. The methodology is demonstrated through the development and implementation of two model discrimination utilities; mutual information and total separation, but it can also be applied more generally if one has different experimental aims. A sequential Monte Carlo algorithm is run for each rival model (in parallel), and provides a convenient estimate of the marginal likelihood (of each model) given the data, which can be used for model comparison and in the evaluation of utility functions. A major benefit of this approach is that it requires very little problem specific tuning and is also computationally efficient when compared to full Markov chain Monte Carlo approaches. This research is motivated by applications in drug development and chemical engineering.

Formato

application/pdf

Identificador

http://eprints.qut.edu.au/53813/

Relação

http://eprints.qut.edu.au/53813/1/smc_continuous.pdf

McGree, James, Drovandi, Christopher C., & Pettitt, Anthony N. (2012) A sequential Monte Carlo approach to the sequential design for discriminating between rival continuous data models. [Working Paper] (Unpublished)

Direitos

Copyright 2012 [please consult the author]

Fonte

Science & Engineering Faculty

Palavras-Chave #010400 STATISTICS #Bayesian sequential design #Continuous response #Model discrimination #Mutual information #Nonlinear models #Particle filter #Total separation
Tipo

Working Paper