Efficient Bayesian Estimation of Multivariate State Space Models

Strickland, Christopher M. and Turner, Ian W. and Denham, Robert and Mengersen, Kerrie L. (2008) Efficient Bayesian Estimation of Multivariate State Space Models.

Full text available as: PDF (Author version) - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

A Bayesian Markov chain Monte Carlo methodology is developed for the estimation of multivariate linear Gaussian state space models. In particular, an efficient simulation smoothing algorithm is proposed that makes use of the univariate representation of the state space model. Substantial gains over existing algorithms in computational efficiency are achieved using the new simulation smoother for the analysis of high dimensional multivariate time series. The methodology is used to analyse a multivariate timeseries dataset of the Normalised Difference Vegetation Index (NDVI), which is a proxy for the level of live vegetation, for a particular grazing property located in Queensland, Australia.

Item Type:	Preprint
Keywords:	Multivariate; State space model; Markov chain Monte Carlo; Kalman filter; Simulation smoother; Univiarate representation; MODIS; Stochastic cycle
Subjects:	230000 Mathematical Sciences > 230100 Mathematics > 230116 Numerical Analysis 230000 Mathematical Sciences > 230200 Statistics > 230204 Applied Statistics 230000 Mathematical Sciences > 230200 Statistics > 230299 Statistics not elsewhere classified
ID Code:	12499
Deposited By:	Strickland, Christopher
Deposited On:	15 February 2008
Copyright Owner:	Copyright 2008 (The authors)
Additional Information:	This research was funded by an ARC linkage grant between QUT and the Department of National Resources and Water