MINIMAX OPTIMAL SEQUENTIAL HYPOTHESIS TESTS FOR MARKOV PROCESSES
Under mild Markov assumptions, sufficient conditions for strict minimax optimality of sequential tests for multiple hypotheses under distributional uncertainty are derived. First, the design of optimal sequential tests for simple hypotheses is revisited, and it is shown that the partial derivatives...
Gespeichert in:
Veröffentlicht in: | The Annals of statistics 2020-10, Vol.48 (5), p.2599-2621 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 2621 |
---|---|
container_issue | 5 |
container_start_page | 2599 |
container_title | The Annals of statistics |
container_volume | 48 |
creator | Fauss, Michael Zoubir, Abdelhak M. Poor, H. Vincent |
description | Under mild Markov assumptions, sufficient conditions for strict minimax optimality of sequential tests for multiple hypotheses under distributional uncertainty are derived. First, the design of optimal sequential tests for simple hypotheses is revisited, and it is shown that the partial derivatives of the corresponding cost function are closely related to the performance metrics of the underlying sequential test. Second, an implicit characterization of the least favorable distributions for a given testing policy is stated. By combining the results on optimal sequential tests and least favorable distributions, sufficient conditions for a sequential test to be minimax optimal under general distributional uncertainties are obtained. The cost function of the minimax optimal test is further identified as a generalized f-dissimilarity and the least favorable distributions as those that are most similar with respect to this dissimilarity. Numerical examples for minimax optimal sequential tests under different uncertainties illustrate the theoretical results. |
doi_str_mv | 10.1214/19-AOS1899 |
format | Article |
fullrecord | <record><control><sourceid>jstor_proqu</sourceid><recordid>TN_cdi_proquest_journals_2451169903</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><jstor_id>27028715</jstor_id><sourcerecordid>27028715</sourcerecordid><originalsourceid>FETCH-LOGICAL-c317t-c945033c2f06941ea6e98dda7261ff6556a044cfcbf101dc61fd6a7b4cbfded03</originalsourceid><addsrcrecordid>eNo9kMFLwzAYxYMoOKcX70LBm1DNlyZpc7OMzBW3ZS6d6Cl0aQMOtTPZDv73Riae3uPx4z14CF0CvgUC9A5EWioNhRBHaECAF2khOD9GA4wFTlnG6Sk6C2GDMWaCZgN0P6vm1ax8SdSijjpNtHxayXldRTt5Xah6InWlk1rqWidjtUxm5fJRPSeLpRpJraU-RyeueQ_dxZ8O0Wos69EknaqHalROU5tBvkutoAxnmSUOc0Gha3gnirZtcsLBOc4YbzCl1tm1AwytjWnLm3xNY9B2Lc6G6PrQu_X9174LO7Pp9_4zThpCGQAXIvYP0c2Bsr4PwXfObP3bR-O_DWDz-5ABYf4eivDVAd6EXe__SZJjUuTAsh9B1Vwt</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2451169903</pqid></control><display><type>article</type><title>MINIMAX OPTIMAL SEQUENTIAL HYPOTHESIS TESTS FOR MARKOV PROCESSES</title><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><source>JSTOR Mathematics & Statistics</source><source>Jstor Complete Legacy</source><source>Project Euclid Complete</source><creator>Fauss, Michael ; Zoubir, Abdelhak M. ; Poor, H. Vincent</creator><creatorcontrib>Fauss, Michael ; Zoubir, Abdelhak M. ; Poor, H. Vincent</creatorcontrib><description>Under mild Markov assumptions, sufficient conditions for strict minimax optimality of sequential tests for multiple hypotheses under distributional uncertainty are derived. First, the design of optimal sequential tests for simple hypotheses is revisited, and it is shown that the partial derivatives of the corresponding cost function are closely related to the performance metrics of the underlying sequential test. Second, an implicit characterization of the least favorable distributions for a given testing policy is stated. By combining the results on optimal sequential tests and least favorable distributions, sufficient conditions for a sequential test to be minimax optimal under general distributional uncertainties are obtained. The cost function of the minimax optimal test is further identified as a generalized f-dissimilarity and the least favorable distributions as those that are most similar with respect to this dissimilarity. Numerical examples for minimax optimal sequential tests under different uncertainties illustrate the theoretical results.</description><identifier>ISSN: 0090-5364</identifier><identifier>EISSN: 2168-8966</identifier><identifier>DOI: 10.1214/19-AOS1899</identifier><language>eng</language><publisher>Hayward: Institute of Mathematical Statistics</publisher><subject>Asymptotic methods ; Cost function ; Costs ; Hypotheses ; Markov analysis ; Markov processes ; Minimax technique ; Numerical analysis ; Optimization ; Performance measurement ; Stochastic models ; Uncertainty</subject><ispartof>The Annals of statistics, 2020-10, Vol.48 (5), p.2599-2621</ispartof><rights>Institute of Mathematical Statistics, 2020</rights><rights>Copyright Institute of Mathematical Statistics Oct 2020</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c317t-c945033c2f06941ea6e98dda7261ff6556a044cfcbf101dc61fd6a7b4cbfded03</citedby><cites>FETCH-LOGICAL-c317t-c945033c2f06941ea6e98dda7261ff6556a044cfcbf101dc61fd6a7b4cbfded03</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.jstor.org/stable/pdf/27028715$$EPDF$$P50$$Gjstor$$H</linktopdf><linktohtml>$$Uhttps://www.jstor.org/stable/27028715$$EHTML$$P50$$Gjstor$$H</linktohtml><link.rule.ids>314,776,780,799,828,27903,27904,57995,57999,58228,58232</link.rule.ids></links><search><creatorcontrib>Fauss, Michael</creatorcontrib><creatorcontrib>Zoubir, Abdelhak M.</creatorcontrib><creatorcontrib>Poor, H. Vincent</creatorcontrib><title>MINIMAX OPTIMAL SEQUENTIAL HYPOTHESIS TESTS FOR MARKOV PROCESSES</title><title>The Annals of statistics</title><description>Under mild Markov assumptions, sufficient conditions for strict minimax optimality of sequential tests for multiple hypotheses under distributional uncertainty are derived. First, the design of optimal sequential tests for simple hypotheses is revisited, and it is shown that the partial derivatives of the corresponding cost function are closely related to the performance metrics of the underlying sequential test. Second, an implicit characterization of the least favorable distributions for a given testing policy is stated. By combining the results on optimal sequential tests and least favorable distributions, sufficient conditions for a sequential test to be minimax optimal under general distributional uncertainties are obtained. The cost function of the minimax optimal test is further identified as a generalized f-dissimilarity and the least favorable distributions as those that are most similar with respect to this dissimilarity. Numerical examples for minimax optimal sequential tests under different uncertainties illustrate the theoretical results.</description><subject>Asymptotic methods</subject><subject>Cost function</subject><subject>Costs</subject><subject>Hypotheses</subject><subject>Markov analysis</subject><subject>Markov processes</subject><subject>Minimax technique</subject><subject>Numerical analysis</subject><subject>Optimization</subject><subject>Performance measurement</subject><subject>Stochastic models</subject><subject>Uncertainty</subject><issn>0090-5364</issn><issn>2168-8966</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><recordid>eNo9kMFLwzAYxYMoOKcX70LBm1DNlyZpc7OMzBW3ZS6d6Cl0aQMOtTPZDv73Riae3uPx4z14CF0CvgUC9A5EWioNhRBHaECAF2khOD9GA4wFTlnG6Sk6C2GDMWaCZgN0P6vm1ax8SdSijjpNtHxayXldRTt5Xah6InWlk1rqWidjtUxm5fJRPSeLpRpJraU-RyeueQ_dxZ8O0Wos69EknaqHalROU5tBvkutoAxnmSUOc0Gha3gnirZtcsLBOc4YbzCl1tm1AwytjWnLm3xNY9B2Lc6G6PrQu_X9174LO7Pp9_4zThpCGQAXIvYP0c2Bsr4PwXfObP3bR-O_DWDz-5ABYf4eivDVAd6EXe__SZJjUuTAsh9B1Vwt</recordid><startdate>20201001</startdate><enddate>20201001</enddate><creator>Fauss, Michael</creator><creator>Zoubir, Abdelhak M.</creator><creator>Poor, H. Vincent</creator><general>Institute of Mathematical Statistics</general><scope>AAYXX</scope><scope>CITATION</scope><scope>JQ2</scope></search><sort><creationdate>20201001</creationdate><title>MINIMAX OPTIMAL SEQUENTIAL HYPOTHESIS TESTS FOR MARKOV PROCESSES</title><author>Fauss, Michael ; Zoubir, Abdelhak M. ; Poor, H. Vincent</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c317t-c945033c2f06941ea6e98dda7261ff6556a044cfcbf101dc61fd6a7b4cbfded03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Asymptotic methods</topic><topic>Cost function</topic><topic>Costs</topic><topic>Hypotheses</topic><topic>Markov analysis</topic><topic>Markov processes</topic><topic>Minimax technique</topic><topic>Numerical analysis</topic><topic>Optimization</topic><topic>Performance measurement</topic><topic>Stochastic models</topic><topic>Uncertainty</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Fauss, Michael</creatorcontrib><creatorcontrib>Zoubir, Abdelhak M.</creatorcontrib><creatorcontrib>Poor, H. Vincent</creatorcontrib><collection>CrossRef</collection><collection>ProQuest Computer Science Collection</collection><jtitle>The Annals of statistics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Fauss, Michael</au><au>Zoubir, Abdelhak M.</au><au>Poor, H. Vincent</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>MINIMAX OPTIMAL SEQUENTIAL HYPOTHESIS TESTS FOR MARKOV PROCESSES</atitle><jtitle>The Annals of statistics</jtitle><date>2020-10-01</date><risdate>2020</risdate><volume>48</volume><issue>5</issue><spage>2599</spage><epage>2621</epage><pages>2599-2621</pages><issn>0090-5364</issn><eissn>2168-8966</eissn><abstract>Under mild Markov assumptions, sufficient conditions for strict minimax optimality of sequential tests for multiple hypotheses under distributional uncertainty are derived. First, the design of optimal sequential tests for simple hypotheses is revisited, and it is shown that the partial derivatives of the corresponding cost function are closely related to the performance metrics of the underlying sequential test. Second, an implicit characterization of the least favorable distributions for a given testing policy is stated. By combining the results on optimal sequential tests and least favorable distributions, sufficient conditions for a sequential test to be minimax optimal under general distributional uncertainties are obtained. The cost function of the minimax optimal test is further identified as a generalized f-dissimilarity and the least favorable distributions as those that are most similar with respect to this dissimilarity. Numerical examples for minimax optimal sequential tests under different uncertainties illustrate the theoretical results.</abstract><cop>Hayward</cop><pub>Institute of Mathematical Statistics</pub><doi>10.1214/19-AOS1899</doi><tpages>23</tpages><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0090-5364 |
ispartof | The Annals of statistics, 2020-10, Vol.48 (5), p.2599-2621 |
issn | 0090-5364 2168-8966 |
language | eng |
recordid | cdi_proquest_journals_2451169903 |
source | Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals; JSTOR Mathematics & Statistics; Jstor Complete Legacy; Project Euclid Complete |
subjects | Asymptotic methods Cost function Costs Hypotheses Markov analysis Markov processes Minimax technique Numerical analysis Optimization Performance measurement Stochastic models Uncertainty |
title | MINIMAX OPTIMAL SEQUENTIAL HYPOTHESIS TESTS FOR MARKOV PROCESSES |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T20%3A48%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-jstor_proqu&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=MINIMAX%20OPTIMAL%20SEQUENTIAL%20HYPOTHESIS%20TESTS%20FOR%20MARKOV%20PROCESSES&rft.jtitle=The%20Annals%20of%20statistics&rft.au=Fauss,%20Michael&rft.date=2020-10-01&rft.volume=48&rft.issue=5&rft.spage=2599&rft.epage=2621&rft.pages=2599-2621&rft.issn=0090-5364&rft.eissn=2168-8966&rft_id=info:doi/10.1214/19-AOS1899&rft_dat=%3Cjstor_proqu%3E27028715%3C/jstor_proqu%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2451169903&rft_id=info:pmid/&rft_jstor_id=27028715&rfr_iscdi=true |