Regularized regression when covariates are linked on a network: the 3CoSE algorithm
Covariates in regressions may be linked to each other on a network. Knowledge of the network structure can be incorporated into regularized regression settings via a network penalty term. However, when it is unknown whether the connection signs in the network are positive (connected covariates reinf...
Gespeichert in:
| Veröffentlicht in: | Journal of applied statistics 2023-02, Vol.50 (3), p.535-554 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 554 |
|---|---|
| container_issue | 3 |
| container_start_page | 535 |
| container_title | Journal of applied statistics |
| container_volume | 50 |
| creator | Weber, Matthias Striaukas, Jonas Schumacher, Martin Binder, Harald |
| description | Covariates in regressions may be linked to each other on a network. Knowledge of the network structure can be incorporated into regularized regression settings via a network penalty term. However, when it is unknown whether the connection signs in the network are positive (connected covariates reinforce each other) or negative (connected covariates repress each other), the connection signs have to be estimated jointly with the covariate coefficients. This can be done with an algorithm iterating a connection sign estimation step and a covariate coefficient estimation step. We develop such an algorithm, called 3CoSE, and show detailed simulation results and an application forecasting event times. The algorithm performs well in a variety of settings. We also briefly describe the publicly available R-package developed for this purpose. |
| doi_str_mv | 10.1080/02664763.2021.1982878 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmed_primary_36819080</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2779341190</sourcerecordid><originalsourceid>FETCH-LOGICAL-c496t-7fdf3a3ab498f9db61bea9ffbc4b2e121728431da7d7eebdba8c59422f4809393</originalsourceid><addsrcrecordid>eNp9kUtv1DAUhS0EokPhJ4AssWGTwY8ktlmgolELSJWQKKwtJ7mecevYxU5alV9fRzOtgAWruzjfPfdxEHpNyZoSSd4T1ra1aPmaEUbXVEkmhXyCVpS3pCINZ0_RamGqBTpCL3K-JIRI2vDn6Ii3kqriskIX32E7e5Pcbxhwgm2CnF0M-HYHAffxpihmgoxNAuxduCpUUQ0OMN3GdPUBTzvAfBMvTrHx25jctBtfomfW-AyvDvUY_Tw7_bH5Up1_-_x18-m86mvVTpWwg-WGm65W0qqha2kHRlnb9XXHgDIqmKw5HYwYBEA3dEb2jaoZs7Ukiit-jD7ufa_nboShhzAl4_V1cqNJdzoap_9WgtvpbbzRSnEimsXg3cEgxV8z5EmPLvfgvQkQ56yZEIrXtLyqoG__QS_jnEI5b6G4UoxQXqhmT_Up5pzAPi5DiV5i0w-x6SU2fYit9L3585LHroecCnCyB1ywMY2m_N4PejJ3PiabTOhd1vz_M-4BtN-oQw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2773992013</pqid></control><display><type>article</type><title>Regularized regression when covariates are linked on a network: the 3CoSE algorithm</title><source>Business Source Complete</source><source>PubMed Central</source><creator>Weber, Matthias ; Striaukas, Jonas ; Schumacher, Martin ; Binder, Harald</creator><creatorcontrib>Weber, Matthias ; Striaukas, Jonas ; Schumacher, Martin ; Binder, Harald</creatorcontrib><description>Covariates in regressions may be linked to each other on a network. Knowledge of the network structure can be incorporated into regularized regression settings via a network penalty term. However, when it is unknown whether the connection signs in the network are positive (connected covariates reinforce each other) or negative (connected covariates repress each other), the connection signs have to be estimated jointly with the covariate coefficients. This can be done with an algorithm iterating a connection sign estimation step and a covariate coefficient estimation step. We develop such an algorithm, called 3CoSE, and show detailed simulation results and an application forecasting event times. The algorithm performs well in a variety of settings. We also briefly describe the publicly available R-package developed for this purpose.</description><identifier>ISSN: 0266-4763</identifier><identifier>EISSN: 1360-0532</identifier><identifier>DOI: 10.1080/02664763.2021.1982878</identifier><identifier>PMID: 36819080</identifier><language>eng</language><publisher>England: Taylor & Francis</publisher><subject>Algorithms ; high-dimensional data ; machine learning ; network penalty ; Regressions on networks ; Statistical methods</subject><ispartof>Journal of applied statistics, 2023-02, Vol.50 (3), p.535-554</ispartof><rights>2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group 2021</rights><rights>2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.</rights><rights>2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This work is licensed under the Creative Commons Attribution – Non-Commercial – No Derivatives License http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group 2021 The Author(s)</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c496t-7fdf3a3ab498f9db61bea9ffbc4b2e121728431da7d7eebdba8c59422f4809393</citedby><cites>FETCH-LOGICAL-c496t-7fdf3a3ab498f9db61bea9ffbc4b2e121728431da7d7eebdba8c59422f4809393</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC9930759/pdf/$$EPDF$$P50$$Gpubmedcentral$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC9930759/$$EHTML$$P50$$Gpubmedcentral$$Hfree_for_read</linktohtml><link.rule.ids>230,314,724,777,781,882,27905,27906,53772,53774</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/36819080$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Weber, Matthias</creatorcontrib><creatorcontrib>Striaukas, Jonas</creatorcontrib><creatorcontrib>Schumacher, Martin</creatorcontrib><creatorcontrib>Binder, Harald</creatorcontrib><title>Regularized regression when covariates are linked on a network: the 3CoSE algorithm</title><title>Journal of applied statistics</title><addtitle>J Appl Stat</addtitle><description>Covariates in regressions may be linked to each other on a network. Knowledge of the network structure can be incorporated into regularized regression settings via a network penalty term. However, when it is unknown whether the connection signs in the network are positive (connected covariates reinforce each other) or negative (connected covariates repress each other), the connection signs have to be estimated jointly with the covariate coefficients. This can be done with an algorithm iterating a connection sign estimation step and a covariate coefficient estimation step. We develop such an algorithm, called 3CoSE, and show detailed simulation results and an application forecasting event times. The algorithm performs well in a variety of settings. We also briefly describe the publicly available R-package developed for this purpose.</description><subject>Algorithms</subject><subject>high-dimensional data</subject><subject>machine learning</subject><subject>network penalty</subject><subject>Regressions on networks</subject><subject>Statistical methods</subject><issn>0266-4763</issn><issn>1360-0532</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>0YH</sourceid><recordid>eNp9kUtv1DAUhS0EokPhJ4AssWGTwY8ktlmgolELSJWQKKwtJ7mecevYxU5alV9fRzOtgAWruzjfPfdxEHpNyZoSSd4T1ra1aPmaEUbXVEkmhXyCVpS3pCINZ0_RamGqBTpCL3K-JIRI2vDn6Ii3kqriskIX32E7e5Pcbxhwgm2CnF0M-HYHAffxpihmgoxNAuxduCpUUQ0OMN3GdPUBTzvAfBMvTrHx25jctBtfomfW-AyvDvUY_Tw7_bH5Up1_-_x18-m86mvVTpWwg-WGm65W0qqha2kHRlnb9XXHgDIqmKw5HYwYBEA3dEb2jaoZs7Ukiit-jD7ufa_nboShhzAl4_V1cqNJdzoap_9WgtvpbbzRSnEimsXg3cEgxV8z5EmPLvfgvQkQ56yZEIrXtLyqoG__QS_jnEI5b6G4UoxQXqhmT_Up5pzAPi5DiV5i0w-x6SU2fYit9L3585LHroecCnCyB1ywMY2m_N4PejJ3PiabTOhd1vz_M-4BtN-oQw</recordid><startdate>20230217</startdate><enddate>20230217</enddate><creator>Weber, Matthias</creator><creator>Striaukas, Jonas</creator><creator>Schumacher, Martin</creator><creator>Binder, Harald</creator><general>Taylor & Francis</general><general>Taylor & Francis Ltd</general><scope>0YH</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>8FD</scope><scope>H8D</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>7X8</scope><scope>5PM</scope></search><sort><creationdate>20230217</creationdate><title>Regularized regression when covariates are linked on a network: the 3CoSE algorithm</title><author>Weber, Matthias ; Striaukas, Jonas ; Schumacher, Martin ; Binder, Harald</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c496t-7fdf3a3ab498f9db61bea9ffbc4b2e121728431da7d7eebdba8c59422f4809393</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Algorithms</topic><topic>high-dimensional data</topic><topic>machine learning</topic><topic>network penalty</topic><topic>Regressions on networks</topic><topic>Statistical methods</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Weber, Matthias</creatorcontrib><creatorcontrib>Striaukas, Jonas</creatorcontrib><creatorcontrib>Schumacher, Martin</creatorcontrib><creatorcontrib>Binder, Harald</creatorcontrib><collection>Taylor & Francis Open Access</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>Aerospace Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of applied statistics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Weber, Matthias</au><au>Striaukas, Jonas</au><au>Schumacher, Martin</au><au>Binder, Harald</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Regularized regression when covariates are linked on a network: the 3CoSE algorithm</atitle><jtitle>Journal of applied statistics</jtitle><addtitle>J Appl Stat</addtitle><date>2023-02-17</date><risdate>2023</risdate><volume>50</volume><issue>3</issue><spage>535</spage><epage>554</epage><pages>535-554</pages><issn>0266-4763</issn><eissn>1360-0532</eissn><abstract>Covariates in regressions may be linked to each other on a network. Knowledge of the network structure can be incorporated into regularized regression settings via a network penalty term. However, when it is unknown whether the connection signs in the network are positive (connected covariates reinforce each other) or negative (connected covariates repress each other), the connection signs have to be estimated jointly with the covariate coefficients. This can be done with an algorithm iterating a connection sign estimation step and a covariate coefficient estimation step. We develop such an algorithm, called 3CoSE, and show detailed simulation results and an application forecasting event times. The algorithm performs well in a variety of settings. We also briefly describe the publicly available R-package developed for this purpose.</abstract><cop>England</cop><pub>Taylor & Francis</pub><pmid>36819080</pmid><doi>10.1080/02664763.2021.1982878</doi><tpages>20</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0266-4763 |
| ispartof | Journal of applied statistics, 2023-02, Vol.50 (3), p.535-554 |
| issn | 0266-4763 1360-0532 |
| language | eng |
| recordid | cdi_pubmed_primary_36819080 |
| source | Business Source Complete; PubMed Central |
| subjects | Algorithms high-dimensional data machine learning network penalty Regressions on networks Statistical methods |
| title | Regularized regression when covariates are linked on a network: the 3CoSE algorithm |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-20T15%3A26%3A34IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Regularized%20regression%20when%20covariates%20are%20linked%20on%20a%20network:%20the%203CoSE%20algorithm&rft.jtitle=Journal%20of%20applied%20statistics&rft.au=Weber,%20Matthias&rft.date=2023-02-17&rft.volume=50&rft.issue=3&rft.spage=535&rft.epage=554&rft.pages=535-554&rft.issn=0266-4763&rft.eissn=1360-0532&rft_id=info:doi/10.1080/02664763.2021.1982878&rft_dat=%3Cproquest_pubme%3E2779341190%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2773992013&rft_id=info:pmid/36819080&rfr_iscdi=true |