A1226
Title: The block-correlated pseudo marginal sampler for state space models
Authors: David Gunawan - University of Wollongong (Australia) [presenting]
Robert Kohn - University of New South Wales (Australia)
Pratiti Chatterjee - University of New South Wales (Australia)
Abstract: Particle Marginal Metropolis-Hastings (PMMH) is a general approach to Bayesian inference when the likelihood is intractable but can be estimated unbiasedly. An efficient PMMH method is developed that scales up better to higher dimensional state vectors than previous approaches. The following innovations achieve the improvement. First, the trimmed mean of the unbiased likelihood estimates of the multiple particle filters is used. Second, a novel block version of PMMH that works with multiple particle filters is proposed. Third, the article develops an efficient auxiliary disturbance particle filter, necessary when the bootstrap disturbance filter is inefficient, but the state transition density cannot be expressed in closed form. Fourth, a novel sorting algorithm, which is as effective as previous approaches but significantly faster than them, is developed to preserve the correlation between the logs of the likelihood estimates at the current and proposed parameter values. The sampler's performance is investigated empirically by applying it to non-linear Dynamic Stochastic General Equilibrium models with relatively high state dimensions and intractable state transition densities and to multivariate stochastic volatility in the mean models.