Statistics Papers

Document Type

Journal Article

Date of this Version

2009

Publication Source

IFAC Proceedings Volumes

Volume

42

Issue

10

Start Page

1591

Last Page

1596

DOI

10.3182/20090706-3-FR-2004.00264

Abstract

Many nonlinear dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We consider two such models: the switching linear dynamical system (SLDS) and the switching vector autoregressive (VAR) process. Our nonparametric Bayesian approach utilizes a hierarchical Dirichlet process prior to learn an unknown number of persistent, smooth dynamical modes. We develop a sampling algorithm that combines a truncated approximation to the Dirichlet process with efficient joint sampling of the mode and state sequences. The utility and flexibility of our model are demonstrated on synthetic data, sequences of dancing honey bees, and the IBOVESPA stock index.

Copyright/Permission Statement

© 2009. This manuscript version is made available under the CC-BY-NC-ND 4.0 license.

Keywords

jump processes, non-parametric identification, stochastic realization, machine learning, dynamic systems, Markov models, state-space models, autoregressive processes

Share

COinS
 

Date Posted: 27 November 2017

This document has been peer reviewed.