Bayesian system identification via Markov chain Monte Carlo techniques

Authors:
Brett Ninness;Soren Henriksen
Affiliations:
School of Electrical Engineering and Computer Science, The University of Newcastle, Australia;School of Electrical Engineering and Computer Science, The University of Newcastle, Australia
Venue:
Automatica (Journal of IFAC)
Year:
2010

Citing 8
Cited 1

System identification: theory for the user

System identification: theory for the user
Identification and application of bounded-parameter models

Automatica (Journal of IFAC)
Learning dynamical systems in a stationary environment

Systems & Control Letters - Special issue: learning theory
The Metropolis Algorithm

Computing in Science and Engineering
Bayesian system identification

Automatica (Journal of IFAC)
Brief Finite sample properties of system identification of ARX models under mixing conditions

Automatica (Journal of IFAC)
Bootstrap-based estimates of uncertainty in subspace identification methods

Automatica (Journal of IFAC)
Guaranteed non-asymptotic confidence regions in system identification

Automatica (Journal of IFAC)

Computational system identification for Bayesian NARMAX modelling

Automatica (Journal of IFAC)

Quantified Score

Hi-index	22.15

Visualization

Abstract

The work here explores new numerical methods for supporting a Bayesian approach to parameter estimation of dynamic systems. This is primarily motivated by the goal of providing accurate quantification of estimation error that is valid for arbitrary, and hence even very short length data records. The main innovation is the employment of the Metropolis-Hastings algorithm to construct an ergodic Markov chain with invariant density equal to the required posterior density. Monte Carlo analysis of samples from this chain then provides a means for efficiently and accurately computing posteriors for model parameters and arbitrary functions of them.