Tempered particle filtering

The accuracy of particle filters for nonlinear state-space models crucially depends on the proposal distribution that mutates time t−1 particle values into time t values. In the widely-used bootstrap particle filter, this distribution is generated by the state-transition equation. While straightforw...

Full description

Saved in:
Bibliographic Details
Published in:Journal of econometrics 2019-05, Vol.210 (1), p.26-44
Main Authors: Herbst, Edward, Schorfheide, Frank
Format: Article
Language:English
Subjects:
Bayesian analysis
DSGE models
Filters
Measurement
Monte Carlo methods
Monte Carlo simulation
Nonlinear filtering
Nonlinear systems
Particle size
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The accuracy of particle filters for nonlinear state-space models crucially depends on the proposal distribution that mutates time t−1 particle values into time t values. In the widely-used bootstrap particle filter, this distribution is generated by the state-transition equation. While straightforward to implement, the practical performance is often poor. We develop a self-tuning particle filter in which the proposal distribution is constructed adaptively through a sequence of Monte Carlo steps. Intuitively, we start from a measurement error distribution with an inflated variance, and then gradually reduce the variance to its nominal level in a sequence of tempering steps. We show that the filter generates an unbiased and consistent approximation of the likelihood function. Holding the run time fixed, our filter is substantially more accurate in two DSGE model applications than the bootstrap particle filter.
ISSN:0304-4076
1872-6895
DOI:10.1016/j.jeconom.2018.11.003