Estimation of predictive performance for test data in applicability domains using y‐randomization
A new measure of predictive performance for objective variables in regression analysis is proposed, enabling the y‐errors of new samples or test samples to be estimated in the applicability domains (ADs) of regression models. The proposed measure, based on y‐randomization, considers chance correlati...
Gespeichert in:
| Veröffentlicht in: | Journal of chemometrics 2019-09, Vol.33 (9), p.n/a |
|---|---|
| 1. Verfasser: | |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | n/a |
|---|---|
| container_issue | 9 |
| container_start_page | |
| container_title | Journal of chemometrics |
| container_volume | 33 |
| creator | Kaneko, Hiromasa |
| description | A new measure of predictive performance for objective variables in regression analysis is proposed, enabling the y‐errors of new samples or test samples to be estimated in the applicability domains (ADs) of regression models. The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples considering the influence of chance correlations in the given dataset on the regression models. Experiments using numerical simulation, quantitative structure‐activity relationship, and quantitative structure‐property relationship datasets confirm that MAECCE can estimate the distribution of y‐errors of new samples in ADs for various training datasets, descriptor sets, and regression analysis methods, enabling chance correlations to be eliminated from data analysis. Python and MATLAB codes for the proposed algorithm are available at https://github.com/hkaneko1985/maecce.
The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples or test samples in the applicability domains (ADs) of regression models, considering the influence of chance correlations in the given dataset on the regression models. |
| doi_str_mv | 10.1002/cem.3171 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2288558727</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2288558727</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3591-1512c77b7afb08b277fca0dc3c93307818e8f608fa412f9e407bb5ec7334a93d3</originalsourceid><addsrcrecordid>eNp1kM1KAzEQx4MoWKvgIwS8eNmaj90mOUqpH1DxouAtZLOJpOxm1yRV1pOP4DP6JKatV08zzPz4z_AD4ByjGUaIXGnTzShm-ABMMBKiwIS_HIIJ4nxeCMrpMTiJcY1Q3tFyAvQyJtep5HoPewuHYBqnk3s3cDDB9qFTXhuYG5hMTLBRSUHnoRqG1mlVu9alETZ9p5yPcBOdf4Xjz9d3UD4P3ecu-BQcWdVGc_ZXp-D5Zvm0uCtWj7f3i-tVoWklcIErTDRjNVO2RrwmjFmtUKOpFpQixjE33M4Rt6rExApTIlbXldGM0lIJ2tApuNjnDqF_2-R35brfBJ9PSkI4ryrOCMvU5Z7SoY8xGCuHkA2EUWIktwplVii3CjNa7NEP15rxX04ulg87_hemF3SR</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2288558727</pqid></control><display><type>article</type><title>Estimation of predictive performance for test data in applicability domains using y‐randomization</title><source>Wiley Online Library Journals Frontfile Complete</source><creator>Kaneko, Hiromasa</creator><creatorcontrib>Kaneko, Hiromasa</creatorcontrib><description>A new measure of predictive performance for objective variables in regression analysis is proposed, enabling the y‐errors of new samples or test samples to be estimated in the applicability domains (ADs) of regression models. The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples considering the influence of chance correlations in the given dataset on the regression models. Experiments using numerical simulation, quantitative structure‐activity relationship, and quantitative structure‐property relationship datasets confirm that MAECCE can estimate the distribution of y‐errors of new samples in ADs for various training datasets, descriptor sets, and regression analysis methods, enabling chance correlations to be eliminated from data analysis. Python and MATLAB codes for the proposed algorithm are available at https://github.com/hkaneko1985/maecce.
The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples or test samples in the applicability domains (ADs) of regression models, considering the influence of chance correlations in the given dataset on the regression models.</description><identifier>ISSN: 0886-9383</identifier><identifier>EISSN: 1099-128X</identifier><identifier>DOI: 10.1002/cem.3171</identifier><language>eng</language><publisher>Chichester: Wiley Subscription Services, Inc</publisher><subject>Algorithms ; applicability domains ; Computer simulation ; Correlation analysis ; Data analysis ; Datasets ; Domains ; Performance prediction ; predictive performance ; QSAR ; QSPR ; Randomization ; regression ; Regression analysis ; Regression models ; Training ; y‐randomization</subject><ispartof>Journal of chemometrics, 2019-09, Vol.33 (9), p.n/a</ispartof><rights>2019 John Wiley & Sons, Ltd.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c3591-1512c77b7afb08b277fca0dc3c93307818e8f608fa412f9e407bb5ec7334a93d3</citedby><cites>FETCH-LOGICAL-c3591-1512c77b7afb08b277fca0dc3c93307818e8f608fa412f9e407bb5ec7334a93d3</cites><orcidid>0000-0001-8367-6476</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1002%2Fcem.3171$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1002%2Fcem.3171$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>314,776,780,1411,27901,27902,45550,45551</link.rule.ids></links><search><creatorcontrib>Kaneko, Hiromasa</creatorcontrib><title>Estimation of predictive performance for test data in applicability domains using y‐randomization</title><title>Journal of chemometrics</title><description>A new measure of predictive performance for objective variables in regression analysis is proposed, enabling the y‐errors of new samples or test samples to be estimated in the applicability domains (ADs) of regression models. The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples considering the influence of chance correlations in the given dataset on the regression models. Experiments using numerical simulation, quantitative structure‐activity relationship, and quantitative structure‐property relationship datasets confirm that MAECCE can estimate the distribution of y‐errors of new samples in ADs for various training datasets, descriptor sets, and regression analysis methods, enabling chance correlations to be eliminated from data analysis. Python and MATLAB codes for the proposed algorithm are available at https://github.com/hkaneko1985/maecce.
The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples or test samples in the applicability domains (ADs) of regression models, considering the influence of chance correlations in the given dataset on the regression models.</description><subject>Algorithms</subject><subject>applicability domains</subject><subject>Computer simulation</subject><subject>Correlation analysis</subject><subject>Data analysis</subject><subject>Datasets</subject><subject>Domains</subject><subject>Performance prediction</subject><subject>predictive performance</subject><subject>QSAR</subject><subject>QSPR</subject><subject>Randomization</subject><subject>regression</subject><subject>Regression analysis</subject><subject>Regression models</subject><subject>Training</subject><subject>y‐randomization</subject><issn>0886-9383</issn><issn>1099-128X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNp1kM1KAzEQx4MoWKvgIwS8eNmaj90mOUqpH1DxouAtZLOJpOxm1yRV1pOP4DP6JKatV08zzPz4z_AD4ByjGUaIXGnTzShm-ABMMBKiwIS_HIIJ4nxeCMrpMTiJcY1Q3tFyAvQyJtep5HoPewuHYBqnk3s3cDDB9qFTXhuYG5hMTLBRSUHnoRqG1mlVu9alETZ9p5yPcBOdf4Xjz9d3UD4P3ecu-BQcWdVGc_ZXp-D5Zvm0uCtWj7f3i-tVoWklcIErTDRjNVO2RrwmjFmtUKOpFpQixjE33M4Rt6rExApTIlbXldGM0lIJ2tApuNjnDqF_2-R35brfBJ9PSkI4ryrOCMvU5Z7SoY8xGCuHkA2EUWIktwplVii3CjNa7NEP15rxX04ulg87_hemF3SR</recordid><startdate>201909</startdate><enddate>201909</enddate><creator>Kaneko, Hiromasa</creator><general>Wiley Subscription Services, Inc</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7U5</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><orcidid>https://orcid.org/0000-0001-8367-6476</orcidid></search><sort><creationdate>201909</creationdate><title>Estimation of predictive performance for test data in applicability domains using y‐randomization</title><author>Kaneko, Hiromasa</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3591-1512c77b7afb08b277fca0dc3c93307818e8f608fa412f9e407bb5ec7334a93d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Algorithms</topic><topic>applicability domains</topic><topic>Computer simulation</topic><topic>Correlation analysis</topic><topic>Data analysis</topic><topic>Datasets</topic><topic>Domains</topic><topic>Performance prediction</topic><topic>predictive performance</topic><topic>QSAR</topic><topic>QSPR</topic><topic>Randomization</topic><topic>regression</topic><topic>Regression analysis</topic><topic>Regression models</topic><topic>Training</topic><topic>y‐randomization</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Kaneko, Hiromasa</creatorcontrib><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research 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><jtitle>Journal of chemometrics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kaneko, Hiromasa</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Estimation of predictive performance for test data in applicability domains using y‐randomization</atitle><jtitle>Journal of chemometrics</jtitle><date>2019-09</date><risdate>2019</risdate><volume>33</volume><issue>9</issue><epage>n/a</epage><issn>0886-9383</issn><eissn>1099-128X</eissn><abstract>A new measure of predictive performance for objective variables in regression analysis is proposed, enabling the y‐errors of new samples or test samples to be estimated in the applicability domains (ADs) of regression models. The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples considering the influence of chance correlations in the given dataset on the regression models. Experiments using numerical simulation, quantitative structure‐activity relationship, and quantitative structure‐property relationship datasets confirm that MAECCE can estimate the distribution of y‐errors of new samples in ADs for various training datasets, descriptor sets, and regression analysis methods, enabling chance correlations to be eliminated from data analysis. Python and MATLAB codes for the proposed algorithm are available at https://github.com/hkaneko1985/maecce.
The proposed measure, based on y‐randomization, considers chance correlations and is calculated using only training data. This chance correlation‐excluded mean absolute error (MAECCE) can estimate the y‐errors of new samples or test samples in the applicability domains (ADs) of regression models, considering the influence of chance correlations in the given dataset on the regression models.</abstract><cop>Chichester</cop><pub>Wiley Subscription Services, Inc</pub><doi>10.1002/cem.3171</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0001-8367-6476</orcidid></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0886-9383 |
| ispartof | Journal of chemometrics, 2019-09, Vol.33 (9), p.n/a |
| issn | 0886-9383 1099-128X |
| language | eng |
| recordid | cdi_proquest_journals_2288558727 |
| source | Wiley Online Library Journals Frontfile Complete |
| subjects | Algorithms applicability domains Computer simulation Correlation analysis Data analysis Datasets Domains Performance prediction predictive performance QSAR QSPR Randomization regression Regression analysis Regression models Training y‐randomization |
| title | Estimation of predictive performance for test data in applicability domains using y‐randomization |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-29T19%3A59%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Estimation%20of%20predictive%20performance%20for%20test%20data%20in%20applicability%20domains%20using%20y%E2%80%90randomization&rft.jtitle=Journal%20of%20chemometrics&rft.au=Kaneko,%20Hiromasa&rft.date=2019-09&rft.volume=33&rft.issue=9&rft.epage=n/a&rft.issn=0886-9383&rft.eissn=1099-128X&rft_id=info:doi/10.1002/cem.3171&rft_dat=%3Cproquest_cross%3E2288558727%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2288558727&rft_id=info:pmid/&rfr_iscdi=true |