Approximating Posterior Predictive Distributions by Averaging Output From Many Particle Filters
This work addresses computational challenges in Bayesian inference for statisticians and machine learning practitioners, but it appears incremental as it builds on existing particle filter methods.
The paper tackles the problem of approximating posterior predictive distributions by introducing the particle swarm filter, a parallel algorithm that averages outputs from many particle filters, and demonstrates its performance with theoretical guarantees and a numerical study on simulated stochastic volatility data.
This paper introduces the {\it particle swarm filter} (not to be confused with particle swarm optimization): a recursive and embarrassingly parallel algorithm that targets an approximation to the sequence of posterior predictive distributions by averaging expectation approximations from many particle filters. A law of large numbers and a central limit theorem are provided, as well as an numerical study of simulated data from a stochastic volatility model.