Adaptive Mixture Modelling Metropolis Methods for Bayesian Analysis of Non-linear State-Space Models.
Data(s) |
01/06/2010
|
---|---|
Formato |
260 - 280 |
Identificador |
http://www.ncbi.nlm.nih.gov/pubmed/20563281 J Comput Graph Stat, 2010, 19 (2), pp. 260 - 280 1061-8600 |
Idioma(s) |
ENG en_US |
Relação |
J Comput Graph Stat Journal of Computational and Graphical Statistics Journal of Computational and Graphical Statistics |
Palavras-Chave | #Bayesian computation #Forward filtering #backward sampling #Regenerating mixture procedure #Smoothing in state-space models #Systems biology |
Tipo |
Journal Article |
Cobertura |
United States |
Resumo |
We describe a strategy for Markov chain Monte Carlo analysis of non-linear, non-Gaussian state-space models involving batch analysis for inference on dynamic, latent state variables and fixed model parameters. The key innovation is a Metropolis-Hastings method for the time series of state variables based on sequential approximation of filtering and smoothing densities using normal mixtures. These mixtures are propagated through the non-linearities using an accurate, local mixture approximation method, and we use a regenerating procedure to deal with potential degeneracy of mixture components. This provides accurate, direct approximations to sequential filtering and retrospective smoothing distributions, and hence a useful construction of global Metropolis proposal distributions for simulation of posteriors for the set of states. This analysis is embedded within a Gibbs sampler to include uncertain fixed parameters. We give an example motivated by an application in systems biology. Supplemental materials provide an example based on a stochastic volatility model as well as MATLAB code. |