Sampling-free linear Bayesian updating of model state and parameters using a square root approach

Authors:
Oliver Pajonk;Bojana V. Rosić;Hermann G. Matthies
Affiliations:
SPT Group GmbH, Weidestraíe 134, 22083 Hamburg, Germany and Institute of Scientific Computing, Technische Universität Braunschweig, Hans-Sommer-Straíe 65, 38106 Braunschweig, German ...;Institute of Scientific Computing, Technische Universität Braunschweig, Hans-Sommer-Straíe 65, 38106 Braunschweig, Germany;Institute of Scientific Computing, Technische Universität Braunschweig, Hans-Sommer-Straíe 65, 38106 Braunschweig, Germany
Venue:
Computers & Geosciences
Year:
2013

Citing 7
Cited 0

A sparse matrix arithmetic based on H-matrices. Part I: introduction to H-matrices

Computing
Optimization by Vector Space Methods

Optimization by Vector Space Methods
Inverse Problem Theory and Methods for Model Parameter Estimation

Inverse Problem Theory and Methods for Model Parameter Estimation
Generalized spectral decomposition for stochastic nonlinear problems

Journal of Computational Physics
A generalized polynomial chaos based ensemble Kalman filter with high accuracy

Journal of Computational Physics
Numerical Methods for Stochastic Computations: A Spectral Method Approach

Numerical Methods for Stochastic Computations: A Spectral Method Approach
Bayesian inference with optimal maps

Journal of Computational Physics

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a sampling-free implementation of a linear Bayesian filter based on a square root formulation. It employs spectral series expansions of the involved random variables, one such example being Wiener's polynomial chaos. The method is compared to several related methods, as well as a full Bayesian update, on a simple scalar example. Additionally it is applied to a combined state and parameter estimation problem for a chaotic system, the well-known Lorenz-63 model. There, we compare it to the ensemble square root filter (EnSRF), which is essentially a probabilistic implementation of the same underlying estimator. The spectral method is found to be more robust than the probabilistic one, especially for variance estimation. This is to be expected due to the sampling-free implementation.