Localization in high-dimensional Monte Carlo filtering

DOI10.1007/978-3-319-54084-9_8zbMATH Open1364.62270arXiv1610.03701OpenAlexW2531380964MaRDI QIDQ5267858FDOQ5267858

Sylvain Robert, H. R. Künsch

Publication date: 13 June 2017

Published in: Springer Proceedings in Mathematics & Statistics (Search for Journal in Brave)

Abstract: The high dimensionality and computational constraints associated with filtering problems in large-scale geophysical applications are particularly challenging for the Particle Filter (PF). Approximate but efficient methods such as the Ensemble Kalman Filter (EnKF) are therefore usually preferred. A key element of these approximate methods is localization, which is in principle a general technique to avoid the curse of dimensionality and consists in limiting the influence of observations to neighboring sites. However, while it works effectively with the EnKF, localization introduces harmful discontinuities in the estimated physical fields when applied blindly to the PF. In the present paper, we explore two possible local algorithms based on the EnKPF, a hybrid method combining the EnKF and the PF. A simulation study in a conjugate normal setup allows to highlight the trade-offs involved when applying localization to PF type of algorithms in the high-dimensional setting. Experiments with the Lorenz96 model demonstrate the ability of the local EnKPF algorithms to perform well even with a small number of particles compared to the problem size.

Full work available at URL: https://arxiv.org/abs/1610.03701

Recommendations

• Can local particle filters beat the curse of dimensionality?
• Filtering with state space localized Kalman gain
• A local ensemble data assimilation algorithm for nonlinear geophysical models
• Curse-of-dimensionality revisited: Collapse of the particle filter in very large scale systems
• Bridging the ensemble Kalman and particle filters

zbMATH Keywords

curse of dimensionalitydata assimilationparticle filterensemble Kalman filterMonte Carlo filteringfiltering in high dimensionlocalized filter algorithms

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Inference from stochastic processes and prediction (62M20) Applications of statistics to environmental and related topics (62P12)

Cites Work

• Sequential Monte Carlo Methods in Practice
• Efficient data assimilation for spatiotemporal chaos: a local ensemble transform Kalman filter
• Filtering via Simulation: Auxiliary Particle Filters
• Can local particle filters beat the curse of dimensionality?
• Recursive Monte Carlo filters: algorithms and theoretical analysis
• Title not available (Why is that?)
• Bridging the ensemble Kalman and particle filters

Uses Software

• EnKF

This page was built for publication: Localization in high-dimensional Monte Carlo filtering

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5267858)