Time series segmentation with shifting means hidden markov models
We present a new family of hidden Markov models and apply these to the segmentation of hydrological and environmental time series. The proposed hidden Markov models have a discrete state space and their structure is inspired from the <i>shifting means</i> models introduced by Chernoff an...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2006-01-01
|
Series: | Nonlinear Processes in Geophysics |
Online Access: | http://www.nonlin-processes-geophys.net/13/339/2006/npg-13-339-2006.pdf |
id |
doaj-c244467c588c421790920ce547f80e44 |
---|---|
record_format |
Article |
spelling |
doaj-c244467c588c421790920ce547f80e442020-11-24T21:38:54ZengCopernicus PublicationsNonlinear Processes in Geophysics1023-58091607-79462006-01-01133339352Time series segmentation with shifting means hidden markov modelsAth. KehagiasV. FortinWe present a new family of hidden Markov models and apply these to the segmentation of hydrological and environmental time series. The proposed hidden Markov models have a discrete state space and their structure is inspired from the <i>shifting means</i> models introduced by Chernoff and Zacks and by Salas and Boes. An estimation method inspired from the EM algorithm is proposed, and we show that it can accurately identify multiple change-points in a time series. We also show that the solution obtained using this algorithm can serve as a starting point for a Monte-Carlo Markov chain Bayesian estimation method, thus reducing the computing time needed for the Markov chain to converge to a stationary distribution.http://www.nonlin-processes-geophys.net/13/339/2006/npg-13-339-2006.pdf |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ath. Kehagias V. Fortin |
spellingShingle |
Ath. Kehagias V. Fortin Time series segmentation with shifting means hidden markov models Nonlinear Processes in Geophysics |
author_facet |
Ath. Kehagias V. Fortin |
author_sort |
Ath. Kehagias |
title |
Time series segmentation with shifting means hidden markov models |
title_short |
Time series segmentation with shifting means hidden markov models |
title_full |
Time series segmentation with shifting means hidden markov models |
title_fullStr |
Time series segmentation with shifting means hidden markov models |
title_full_unstemmed |
Time series segmentation with shifting means hidden markov models |
title_sort |
time series segmentation with shifting means hidden markov models |
publisher |
Copernicus Publications |
series |
Nonlinear Processes in Geophysics |
issn |
1023-5809 1607-7946 |
publishDate |
2006-01-01 |
description |
We present a new family of hidden Markov models and apply these to the segmentation of hydrological and environmental time series. The proposed hidden Markov models have a discrete state space and their structure is inspired from the <i>shifting means</i> models introduced by Chernoff and Zacks and by Salas and Boes. An estimation method inspired from the EM algorithm is proposed, and we show that it can accurately identify multiple change-points in a time series. We also show that the solution obtained using this algorithm can serve as a starting point for a Monte-Carlo Markov chain Bayesian estimation method, thus reducing the computing time needed for the Markov chain to converge to a stationary distribution. |
url |
http://www.nonlin-processes-geophys.net/13/339/2006/npg-13-339-2006.pdf |
work_keys_str_mv |
AT athkehagias timeseriessegmentationwithshiftingmeanshiddenmarkovmodels AT vfortin timeseriessegmentationwithshiftingmeanshiddenmarkovmodels |
_version_ |
1725933902283931648 |