Scalable structural break detection

[Display omitted] ► The context of the paper is the analysis of large volume of non-stationary data. ► We fit a piecewise AR model to the analysed time series using the MDL principle. ► The existing method AutoPARM scales inefficiently with the data volume. ► We propose an alternative optimisation s...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied soft computing 2012-11, Vol.12 (11), p.3408-3420
Hauptverfasser: Éltetö, T., Hansen, N., Germain-Renaud, C., Bondon, P.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 3420
container_issue 11
container_start_page 3408
container_title Applied soft computing
container_volume 12
creator Éltetö, T.
Hansen, N.
Germain-Renaud, C.
Bondon, P.
description [Display omitted] ► The context of the paper is the analysis of large volume of non-stationary data. ► We fit a piecewise AR model to the analysed time series using the MDL principle. ► The existing method AutoPARM scales inefficiently with the data volume. ► We propose an alternative optimisation strategy using CMA-ES for the fitting. ► Our method achieves at least one order of magnitude performance improvement. This paper deals with a statistical model fitting procedure for non-stationary time series. This procedure selects the parameters of a piecewise autoregressive model using the Minimum Description Length principle. The existing chromosome representation of the piecewise autoregressive model and its corresponding optimisation algorithm are improved. First, we show that our proposed chromosome representation better captures the intrinsic properties of the piecewise autoregressive model. Second, we apply an optimisation algorithm, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), with which our setup converges faster to the optimal fit. Our proposed method achieves at least one order of magnitude performance improvement compared to the existing solution.
doi_str_mv 10.1016/j.asoc.2012.06.002
format Article
fullrecord <record><control><sourceid>elsevier_hal_p</sourceid><recordid>TN_cdi_hal_primary_oai_HAL_hal_00711843v1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S1568494612002694</els_id><sourcerecordid>S1568494612002694</sourcerecordid><originalsourceid>FETCH-LOGICAL-c334t-4a28e6304422de176a53271a43c4c586b80b58fc3ef7225cd9ad39565240edca3</originalsourceid><addsrcrecordid>eNp9kE1Lw0AQhhdRsFb_gKeCJw-Js5_ZgJdS_IKCB_W8THYnuDU2spsW_PcmVDx6mmF4n4H3YeySQ8mBm5tNibn3pQAuSjAlgDhiM24rUdTG8uNx18YWqlbmlJ3lvIERqoWdsasXjx02HS3ykHZ-2CXsFk0i_FgEGsgPsd-es5MWu0wXv3PO3u7vXlePxfr54Wm1XBdeSjUUCoUlI0EpIQLxyqCWouKopFdeW9NYaLRtvaS2EkL7UGOQtTZaKKDgUc7Z9eHvO3buK8VPTN-ux-gel2s33QAqzq2Sez5mxSHrU59zovYP4OAmJW7jJiVuUuLAjKwYodsDRGOLfaTkso-09RRiGqu60Mf_8B8T-WgA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Scalable structural break detection</title><source>Access via ScienceDirect (Elsevier)</source><creator>Éltetö, T. ; Hansen, N. ; Germain-Renaud, C. ; Bondon, P.</creator><creatorcontrib>Éltetö, T. ; Hansen, N. ; Germain-Renaud, C. ; Bondon, P.</creatorcontrib><description>[Display omitted] ► The context of the paper is the analysis of large volume of non-stationary data. ► We fit a piecewise AR model to the analysed time series using the MDL principle. ► The existing method AutoPARM scales inefficiently with the data volume. ► We propose an alternative optimisation strategy using CMA-ES for the fitting. ► Our method achieves at least one order of magnitude performance improvement. This paper deals with a statistical model fitting procedure for non-stationary time series. This procedure selects the parameters of a piecewise autoregressive model using the Minimum Description Length principle. The existing chromosome representation of the piecewise autoregressive model and its corresponding optimisation algorithm are improved. First, we show that our proposed chromosome representation better captures the intrinsic properties of the piecewise autoregressive model. Second, we apply an optimisation algorithm, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), with which our setup converges faster to the optimal fit. Our proposed method achieves at least one order of magnitude performance improvement compared to the existing solution.</description><identifier>ISSN: 1568-4946</identifier><identifier>EISSN: 1872-9681</identifier><identifier>DOI: 10.1016/j.asoc.2012.06.002</identifier><language>eng</language><publisher>Elsevier B.V</publisher><subject>Computer Science ; Covariance Matrix Adaptation ; Evolution Strategy ; Machine Learning ; Minimum Description Length principle</subject><ispartof>Applied soft computing, 2012-11, Vol.12 (11), p.3408-3420</ispartof><rights>2012 Elsevier B.V.</rights><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c334t-4a28e6304422de176a53271a43c4c586b80b58fc3ef7225cd9ad39565240edca3</citedby><cites>FETCH-LOGICAL-c334t-4a28e6304422de176a53271a43c4c586b80b58fc3ef7225cd9ad39565240edca3</cites><orcidid>0000-0002-5158-7337 ; 0000-0001-7788-4906</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.asoc.2012.06.002$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>230,314,780,784,885,3550,27924,27925,45995</link.rule.ids><backlink>$$Uhttps://inria.hal.science/hal-00711843$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Éltetö, T.</creatorcontrib><creatorcontrib>Hansen, N.</creatorcontrib><creatorcontrib>Germain-Renaud, C.</creatorcontrib><creatorcontrib>Bondon, P.</creatorcontrib><title>Scalable structural break detection</title><title>Applied soft computing</title><description>[Display omitted] ► The context of the paper is the analysis of large volume of non-stationary data. ► We fit a piecewise AR model to the analysed time series using the MDL principle. ► The existing method AutoPARM scales inefficiently with the data volume. ► We propose an alternative optimisation strategy using CMA-ES for the fitting. ► Our method achieves at least one order of magnitude performance improvement. This paper deals with a statistical model fitting procedure for non-stationary time series. This procedure selects the parameters of a piecewise autoregressive model using the Minimum Description Length principle. The existing chromosome representation of the piecewise autoregressive model and its corresponding optimisation algorithm are improved. First, we show that our proposed chromosome representation better captures the intrinsic properties of the piecewise autoregressive model. Second, we apply an optimisation algorithm, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), with which our setup converges faster to the optimal fit. Our proposed method achieves at least one order of magnitude performance improvement compared to the existing solution.</description><subject>Computer Science</subject><subject>Covariance Matrix Adaptation</subject><subject>Evolution Strategy</subject><subject>Machine Learning</subject><subject>Minimum Description Length principle</subject><issn>1568-4946</issn><issn>1872-9681</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><recordid>eNp9kE1Lw0AQhhdRsFb_gKeCJw-Js5_ZgJdS_IKCB_W8THYnuDU2spsW_PcmVDx6mmF4n4H3YeySQ8mBm5tNibn3pQAuSjAlgDhiM24rUdTG8uNx18YWqlbmlJ3lvIERqoWdsasXjx02HS3ykHZ-2CXsFk0i_FgEGsgPsd-es5MWu0wXv3PO3u7vXlePxfr54Wm1XBdeSjUUCoUlI0EpIQLxyqCWouKopFdeW9NYaLRtvaS2EkL7UGOQtTZaKKDgUc7Z9eHvO3buK8VPTN-ux-gel2s33QAqzq2Sez5mxSHrU59zovYP4OAmJW7jJiVuUuLAjKwYodsDRGOLfaTkso-09RRiGqu60Mf_8B8T-WgA</recordid><startdate>20121101</startdate><enddate>20121101</enddate><creator>Éltetö, T.</creator><creator>Hansen, N.</creator><creator>Germain-Renaud, C.</creator><creator>Bondon, P.</creator><general>Elsevier B.V</general><general>Elsevier</general><scope>AAYXX</scope><scope>CITATION</scope><scope>1XC</scope><scope>VOOES</scope><orcidid>https://orcid.org/0000-0002-5158-7337</orcidid><orcidid>https://orcid.org/0000-0001-7788-4906</orcidid></search><sort><creationdate>20121101</creationdate><title>Scalable structural break detection</title><author>Éltetö, T. ; Hansen, N. ; Germain-Renaud, C. ; Bondon, P.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c334t-4a28e6304422de176a53271a43c4c586b80b58fc3ef7225cd9ad39565240edca3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Computer Science</topic><topic>Covariance Matrix Adaptation</topic><topic>Evolution Strategy</topic><topic>Machine Learning</topic><topic>Minimum Description Length principle</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Éltetö, T.</creatorcontrib><creatorcontrib>Hansen, N.</creatorcontrib><creatorcontrib>Germain-Renaud, C.</creatorcontrib><creatorcontrib>Bondon, P.</creatorcontrib><collection>CrossRef</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>Hyper Article en Ligne (HAL) (Open Access)</collection><jtitle>Applied soft computing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Éltetö, T.</au><au>Hansen, N.</au><au>Germain-Renaud, C.</au><au>Bondon, P.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Scalable structural break detection</atitle><jtitle>Applied soft computing</jtitle><date>2012-11-01</date><risdate>2012</risdate><volume>12</volume><issue>11</issue><spage>3408</spage><epage>3420</epage><pages>3408-3420</pages><issn>1568-4946</issn><eissn>1872-9681</eissn><abstract>[Display omitted] ► The context of the paper is the analysis of large volume of non-stationary data. ► We fit a piecewise AR model to the analysed time series using the MDL principle. ► The existing method AutoPARM scales inefficiently with the data volume. ► We propose an alternative optimisation strategy using CMA-ES for the fitting. ► Our method achieves at least one order of magnitude performance improvement. This paper deals with a statistical model fitting procedure for non-stationary time series. This procedure selects the parameters of a piecewise autoregressive model using the Minimum Description Length principle. The existing chromosome representation of the piecewise autoregressive model and its corresponding optimisation algorithm are improved. First, we show that our proposed chromosome representation better captures the intrinsic properties of the piecewise autoregressive model. Second, we apply an optimisation algorithm, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), with which our setup converges faster to the optimal fit. Our proposed method achieves at least one order of magnitude performance improvement compared to the existing solution.</abstract><pub>Elsevier B.V</pub><doi>10.1016/j.asoc.2012.06.002</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0002-5158-7337</orcidid><orcidid>https://orcid.org/0000-0001-7788-4906</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1568-4946
ispartof Applied soft computing, 2012-11, Vol.12 (11), p.3408-3420
issn 1568-4946
1872-9681
language eng
recordid cdi_hal_primary_oai_HAL_hal_00711843v1
source Access via ScienceDirect (Elsevier)
subjects Computer Science
Covariance Matrix Adaptation
Evolution Strategy
Machine Learning
Minimum Description Length principle
title Scalable structural break detection
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T15%3A20%3A25IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-elsevier_hal_p&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Scalable%20structural%20break%20detection&rft.jtitle=Applied%20soft%20computing&rft.au=%C3%89ltet%C3%B6,%20T.&rft.date=2012-11-01&rft.volume=12&rft.issue=11&rft.spage=3408&rft.epage=3420&rft.pages=3408-3420&rft.issn=1568-4946&rft.eissn=1872-9681&rft_id=info:doi/10.1016/j.asoc.2012.06.002&rft_dat=%3Celsevier_hal_p%3ES1568494612002694%3C/elsevier_hal_p%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_els_id=S1568494612002694&rfr_iscdi=true