Clustering time series, subspace identification and cepstral distances

In this paper a methodology to cluster time series based on measurement data is described. In particular, we propose a distance for stochastic models based on the concept of subspace angles within a model and between two models. This distance is used to obtain a clustering over the set of time serie...

Full description

Saved in:
Bibliographic Details
Published in:Communications in information and systems 2005, Vol.5 (1), p.69-96
Main Authors: Boets, Jeroen, Cock, K. de, M., Espinoza, Moor, B. de
Format: Article
Language:English
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper a methodology to cluster time series based on measurement data is described. In particular, we propose a distance for stochastic models based on the concept of subspace angles within a model and between two models. This distance is used to obtain a clustering over the set of time series. We show how it is related to the mutual information of the past and the future output processes, and to a previously defined cepstral distance. Finally, the methodology is applied to the clustering of time series of power consumption within the Belgian electricity grid.
ISSN:1526-7555
2163-4548
DOI:10.4310/CIS.2005.v5.n1.a3