Necessary and sufficient conditions for causal feature selection in time series with latent common causes
We study the identification of direct and indirect causes on time series and provide conditions in the presence of latent variables, which we prove to be necessary and sufficient under some graph constraints. Our theoretical results and estimation algorithms require two conditional independence test...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | Mastakouri, Atalanti A Schölkopf, Bernhard Janzing, Dominik |
| description | We study the identification of direct and indirect causes on time series and
provide conditions in the presence of latent variables, which we prove to be
necessary and sufficient under some graph constraints. Our theoretical results
and estimation algorithms require two conditional independence tests for each
observed candidate time series to determine whether or not it is a cause of an
observed target time series. We provide experimental results in simulations, as
well as real data. Our results show that our method leads to very low false
positives and relatively low false negative rates, outperforming the widely
used Granger causality. |
| doi_str_mv | 10.48550/arxiv.2005.08543 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2005_08543</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2005_08543</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-839e1cca895f3b661cef1314467a9447153635b16c0644690d318d5502e518eb3</originalsourceid><addsrcrecordid>eNotz8lqwzAQBmBdeihpH6Cn6gXsStFi-RhCNwjNJXczlkd0wEuQ5C5vXzvpaeCfmR8-xh6kKLUzRjxB_KGvciuEKYUzWt0y-kCPKUH85TB2PM0hkCccM_fT2FGmaUw8TJF7mBP0PCDkOSJP2KNft5xGnmlYk0iY-DflT95DvlYMw3KxvmK6YzcB-oT3_3PDTi_Pp_1bcTi-vu93hwJspQqnapTeg6tNUK210mOQSmptK6i1rqRRVplWWi_sEtaiU9J1i22LRjps1YY9Xmsv1uYcaVhwzWpuLmb1B7AAUns</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Necessary and sufficient conditions for causal feature selection in time series with latent common causes</title><source>arXiv.org</source><creator>Mastakouri, Atalanti A ; Schölkopf, Bernhard ; Janzing, Dominik</creator><creatorcontrib>Mastakouri, Atalanti A ; Schölkopf, Bernhard ; Janzing, Dominik</creatorcontrib><description>We study the identification of direct and indirect causes on time series and
provide conditions in the presence of latent variables, which we prove to be
necessary and sufficient under some graph constraints. Our theoretical results
and estimation algorithms require two conditional independence tests for each
observed candidate time series to determine whether or not it is a cause of an
observed target time series. We provide experimental results in simulations, as
well as real data. Our results show that our method leads to very low false
positives and relatively low false negative rates, outperforming the widely
used Granger causality.</description><identifier>DOI: 10.48550/arxiv.2005.08543</identifier><language>eng</language><subject>Statistics - Machine Learning ; Statistics - Methodology</subject><creationdate>2020-05</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,777,882</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2005.08543$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2005.08543$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Mastakouri, Atalanti A</creatorcontrib><creatorcontrib>Schölkopf, Bernhard</creatorcontrib><creatorcontrib>Janzing, Dominik</creatorcontrib><title>Necessary and sufficient conditions for causal feature selection in time series with latent common causes</title><description>We study the identification of direct and indirect causes on time series and
provide conditions in the presence of latent variables, which we prove to be
necessary and sufficient under some graph constraints. Our theoretical results
and estimation algorithms require two conditional independence tests for each
observed candidate time series to determine whether or not it is a cause of an
observed target time series. We provide experimental results in simulations, as
well as real data. Our results show that our method leads to very low false
positives and relatively low false negative rates, outperforming the widely
used Granger causality.</description><subject>Statistics - Machine Learning</subject><subject>Statistics - Methodology</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz8lqwzAQBmBdeihpH6Cn6gXsStFi-RhCNwjNJXczlkd0wEuQ5C5vXzvpaeCfmR8-xh6kKLUzRjxB_KGvciuEKYUzWt0y-kCPKUH85TB2PM0hkCccM_fT2FGmaUw8TJF7mBP0PCDkOSJP2KNft5xGnmlYk0iY-DflT95DvlYMw3KxvmK6YzcB-oT3_3PDTi_Pp_1bcTi-vu93hwJspQqnapTeg6tNUK210mOQSmptK6i1rqRRVplWWi_sEtaiU9J1i22LRjps1YY9Xmsv1uYcaVhwzWpuLmb1B7AAUns</recordid><startdate>20200518</startdate><enddate>20200518</enddate><creator>Mastakouri, Atalanti A</creator><creator>Schölkopf, Bernhard</creator><creator>Janzing, Dominik</creator><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20200518</creationdate><title>Necessary and sufficient conditions for causal feature selection in time series with latent common causes</title><author>Mastakouri, Atalanti A ; Schölkopf, Bernhard ; Janzing, Dominik</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-839e1cca895f3b661cef1314467a9447153635b16c0644690d318d5502e518eb3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Statistics - Machine Learning</topic><topic>Statistics - Methodology</topic><toplevel>online_resources</toplevel><creatorcontrib>Mastakouri, Atalanti A</creatorcontrib><creatorcontrib>Schölkopf, Bernhard</creatorcontrib><creatorcontrib>Janzing, Dominik</creatorcontrib><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Mastakouri, Atalanti A</au><au>Schölkopf, Bernhard</au><au>Janzing, Dominik</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Necessary and sufficient conditions for causal feature selection in time series with latent common causes</atitle><date>2020-05-18</date><risdate>2020</risdate><abstract>We study the identification of direct and indirect causes on time series and
provide conditions in the presence of latent variables, which we prove to be
necessary and sufficient under some graph constraints. Our theoretical results
and estimation algorithms require two conditional independence tests for each
observed candidate time series to determine whether or not it is a cause of an
observed target time series. We provide experimental results in simulations, as
well as real data. Our results show that our method leads to very low false
positives and relatively low false negative rates, outperforming the widely
used Granger causality.</abstract><doi>10.48550/arxiv.2005.08543</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.2005.08543 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_2005_08543 |
| source | arXiv.org |
| subjects | Statistics - Machine Learning Statistics - Methodology |
| title | Necessary and sufficient conditions for causal feature selection in time series with latent common causes |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T16%3A23%3A59IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Necessary%20and%20sufficient%20conditions%20for%20causal%20feature%20selection%20in%20time%20series%20with%20latent%20common%20causes&rft.au=Mastakouri,%20Atalanti%20A&rft.date=2020-05-18&rft_id=info:doi/10.48550/arxiv.2005.08543&rft_dat=%3Carxiv_GOX%3E2005_08543%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |