Multiple testing of composite null hypotheses for discrete data using randomized p‐values

P‐values that are derived from continuously distributed test statistics are typically uniformly distributed on (0,1) under least favorable parameter configurations (LFCs) in the null hypothesis. Conservativeness of a p‐value P (meaning that P is under the null hypothesis stochastically larger than u...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Biometrical journal 2024-01, Vol.66 (1), p.e2300077-n/a
Hauptverfasser: Ochieng, Daniel, Hoang, Anh‐Tuan, Dickhaus, Thorsten
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 1
container_start_page e2300077
container_title Biometrical journal
container_volume 66
creator Ochieng, Daniel
Hoang, Anh‐Tuan
Dickhaus, Thorsten
description P‐values that are derived from continuously distributed test statistics are typically uniformly distributed on (0,1) under least favorable parameter configurations (LFCs) in the null hypothesis. Conservativeness of a p‐value P (meaning that P is under the null hypothesis stochastically larger than uniform on (0,1)) can occur if the test statistic from which P is derived is discrete, or if the true parameter value under the null is not an LFC. To deal with both of these sources of conservativeness, we present two approaches utilizing randomized p‐values. We illustrate their effectiveness for testing a composite null hypothesis under a binomial model. We also give an example of how the proposed p‐values can be used to test a composite null in group testing designs. We find that the proposed randomized p‐values are less conservative compared to nonrandomized p‐values under the null hypothesis, but that they are stochastically not smaller under the alternative. The problem of establishing the validity of randomized p‐values has received attention in previous literature. We show that our proposed randomized p‐values are valid under various discrete statistical models, which are such that the distribution of the corresponding test statistic belongs to an exponential family. The behavior of the power function for the tests based on the proposed randomized p‐values as a function of the sample size is also investigated. Simulations and a real data example are used to compare the different considered p‐values.
doi_str_mv 10.1002/bimj.202300077
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2880105322</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2919357870</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3636-f9ab3b3fc265f930054339314add87a66ff343855e81441366c591584f0dbb283</originalsourceid><addsrcrecordid>eNqFkLFS3DAQhjWZMOGAtCkzmklD40PSSrZcJjdALnMMDVQUGtmWgm9ky5FsmEuVR-AZeRJ0c3BFGqot9tt_dz-EvlAyp4Sws6rt1nNGGBBCiuIDmlHBaMYJ5B_RjACDDCQvDtFRjOuElISzT-gQCikKATBDd1eTG9vBGTyaOLb9b-wtrn03-NiOBveTc_h-M_jx3kQTsfUBN22sg0nNRo8aT3E7FHTf-K79axo8PP97etBuMvEEHVjtovn8Wo_R7cX5zeJntrq-XC6-r7IacsgzW-oKKrA1y4Ut0yOCA5RAuW4aWeg8txY4SCGMpJxTyPNalFRIbklTVUzCMTrd5Q7B_0l7R9WlE41zujd-iopJSSgRwFhCv_2Hrv0U-nSdYiUtQRSyIIma76g6-BiDsWoIbafDRlGittrVVrvaa08DX19jp6ozzR5_85wAvgMeW2c278SpH8urX0wkOS-A7Y4X</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2919357870</pqid></control><display><type>article</type><title>Multiple testing of composite null hypotheses for discrete data using randomized p‐values</title><source>Wiley Online Library All Journals</source><creator>Ochieng, Daniel ; Hoang, Anh‐Tuan ; Dickhaus, Thorsten</creator><creatorcontrib>Ochieng, Daniel ; Hoang, Anh‐Tuan ; Dickhaus, Thorsten</creatorcontrib><description>P‐values that are derived from continuously distributed test statistics are typically uniformly distributed on (0,1) under least favorable parameter configurations (LFCs) in the null hypothesis. Conservativeness of a p‐value P (meaning that P is under the null hypothesis stochastically larger than uniform on (0,1)) can occur if the test statistic from which P is derived is discrete, or if the true parameter value under the null is not an LFC. To deal with both of these sources of conservativeness, we present two approaches utilizing randomized p‐values. We illustrate their effectiveness for testing a composite null hypothesis under a binomial model. We also give an example of how the proposed p‐values can be used to test a composite null in group testing designs. We find that the proposed randomized p‐values are less conservative compared to nonrandomized p‐values under the null hypothesis, but that they are stochastically not smaller under the alternative. The problem of establishing the validity of randomized p‐values has received attention in previous literature. We show that our proposed randomized p‐values are valid under various discrete statistical models, which are such that the distribution of the corresponding test statistic belongs to an exponential family. The behavior of the power function for the tests based on the proposed randomized p‐values as a function of the sample size is also investigated. Simulations and a real data example are used to compare the different considered p‐values.</description><identifier>ISSN: 0323-3847</identifier><identifier>EISSN: 1521-4036</identifier><identifier>DOI: 10.1002/bimj.202300077</identifier><identifier>PMID: 37857533</identifier><language>eng</language><publisher>Germany: Wiley - VCH Verlag GmbH &amp; Co. KGaA</publisher><subject>conservative tests ; discretely distributed test statistics ; group testing ; Hypotheses ; Mathematical models ; multiple comparisons ; Null hypothesis ; Parameters ; randomized tests ; Statistical analysis ; Statistical methods ; Statistical models ; Statistical tests</subject><ispartof>Biometrical journal, 2024-01, Vol.66 (1), p.e2300077-n/a</ispartof><rights>2023 The Authors. published by Wiley‐VCH GmbH.</rights><rights>2023 The Authors. Biometrical Journal published by Wiley-VCH GmbH.</rights><rights>2023. This article is published under 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><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c3636-f9ab3b3fc265f930054339314add87a66ff343855e81441366c591584f0dbb283</cites><orcidid>0000-0003-3084-3036</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%2Fbimj.202300077$$EPDF$$P50$$Gwiley$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1002%2Fbimj.202300077$$EHTML$$P50$$Gwiley$$Hfree_for_read</linktohtml><link.rule.ids>315,781,785,1418,27929,27930,45579,45580</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/37857533$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Ochieng, Daniel</creatorcontrib><creatorcontrib>Hoang, Anh‐Tuan</creatorcontrib><creatorcontrib>Dickhaus, Thorsten</creatorcontrib><title>Multiple testing of composite null hypotheses for discrete data using randomized p‐values</title><title>Biometrical journal</title><addtitle>Biom J</addtitle><description>P‐values that are derived from continuously distributed test statistics are typically uniformly distributed on (0,1) under least favorable parameter configurations (LFCs) in the null hypothesis. Conservativeness of a p‐value P (meaning that P is under the null hypothesis stochastically larger than uniform on (0,1)) can occur if the test statistic from which P is derived is discrete, or if the true parameter value under the null is not an LFC. To deal with both of these sources of conservativeness, we present two approaches utilizing randomized p‐values. We illustrate their effectiveness for testing a composite null hypothesis under a binomial model. We also give an example of how the proposed p‐values can be used to test a composite null in group testing designs. We find that the proposed randomized p‐values are less conservative compared to nonrandomized p‐values under the null hypothesis, but that they are stochastically not smaller under the alternative. The problem of establishing the validity of randomized p‐values has received attention in previous literature. We show that our proposed randomized p‐values are valid under various discrete statistical models, which are such that the distribution of the corresponding test statistic belongs to an exponential family. The behavior of the power function for the tests based on the proposed randomized p‐values as a function of the sample size is also investigated. Simulations and a real data example are used to compare the different considered p‐values.</description><subject>conservative tests</subject><subject>discretely distributed test statistics</subject><subject>group testing</subject><subject>Hypotheses</subject><subject>Mathematical models</subject><subject>multiple comparisons</subject><subject>Null hypothesis</subject><subject>Parameters</subject><subject>randomized tests</subject><subject>Statistical analysis</subject><subject>Statistical methods</subject><subject>Statistical models</subject><subject>Statistical tests</subject><issn>0323-3847</issn><issn>1521-4036</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>24P</sourceid><sourceid>WIN</sourceid><recordid>eNqFkLFS3DAQhjWZMOGAtCkzmklD40PSSrZcJjdALnMMDVQUGtmWgm9ky5FsmEuVR-AZeRJ0c3BFGqot9tt_dz-EvlAyp4Sws6rt1nNGGBBCiuIDmlHBaMYJ5B_RjACDDCQvDtFRjOuElISzT-gQCikKATBDd1eTG9vBGTyaOLb9b-wtrn03-NiOBveTc_h-M_jx3kQTsfUBN22sg0nNRo8aT3E7FHTf-K79axo8PP97etBuMvEEHVjtovn8Wo_R7cX5zeJntrq-XC6-r7IacsgzW-oKKrA1y4Ut0yOCA5RAuW4aWeg8txY4SCGMpJxTyPNalFRIbklTVUzCMTrd5Q7B_0l7R9WlE41zujd-iopJSSgRwFhCv_2Hrv0U-nSdYiUtQRSyIIma76g6-BiDsWoIbafDRlGittrVVrvaa08DX19jp6ozzR5_85wAvgMeW2c278SpH8urX0wkOS-A7Y4X</recordid><startdate>202401</startdate><enddate>202401</enddate><creator>Ochieng, Daniel</creator><creator>Hoang, Anh‐Tuan</creator><creator>Dickhaus, Thorsten</creator><general>Wiley - VCH Verlag GmbH &amp; Co. KGaA</general><scope>24P</scope><scope>WIN</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7QO</scope><scope>8FD</scope><scope>FR3</scope><scope>K9.</scope><scope>P64</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0003-3084-3036</orcidid></search><sort><creationdate>202401</creationdate><title>Multiple testing of composite null hypotheses for discrete data using randomized p‐values</title><author>Ochieng, Daniel ; Hoang, Anh‐Tuan ; Dickhaus, Thorsten</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3636-f9ab3b3fc265f930054339314add87a66ff343855e81441366c591584f0dbb283</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>conservative tests</topic><topic>discretely distributed test statistics</topic><topic>group testing</topic><topic>Hypotheses</topic><topic>Mathematical models</topic><topic>multiple comparisons</topic><topic>Null hypothesis</topic><topic>Parameters</topic><topic>randomized tests</topic><topic>Statistical analysis</topic><topic>Statistical methods</topic><topic>Statistical models</topic><topic>Statistical tests</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Ochieng, Daniel</creatorcontrib><creatorcontrib>Hoang, Anh‐Tuan</creatorcontrib><creatorcontrib>Dickhaus, Thorsten</creatorcontrib><collection>Wiley-Blackwell Open Access Titles</collection><collection>Wiley Free Content</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Biotechnology Research Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>ProQuest Health &amp; Medical Complete (Alumni)</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>MEDLINE - Academic</collection><jtitle>Biometrical journal</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ochieng, Daniel</au><au>Hoang, Anh‐Tuan</au><au>Dickhaus, Thorsten</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Multiple testing of composite null hypotheses for discrete data using randomized p‐values</atitle><jtitle>Biometrical journal</jtitle><addtitle>Biom J</addtitle><date>2024-01</date><risdate>2024</risdate><volume>66</volume><issue>1</issue><spage>e2300077</spage><epage>n/a</epage><pages>e2300077-n/a</pages><issn>0323-3847</issn><eissn>1521-4036</eissn><abstract>P‐values that are derived from continuously distributed test statistics are typically uniformly distributed on (0,1) under least favorable parameter configurations (LFCs) in the null hypothesis. Conservativeness of a p‐value P (meaning that P is under the null hypothesis stochastically larger than uniform on (0,1)) can occur if the test statistic from which P is derived is discrete, or if the true parameter value under the null is not an LFC. To deal with both of these sources of conservativeness, we present two approaches utilizing randomized p‐values. We illustrate their effectiveness for testing a composite null hypothesis under a binomial model. We also give an example of how the proposed p‐values can be used to test a composite null in group testing designs. We find that the proposed randomized p‐values are less conservative compared to nonrandomized p‐values under the null hypothesis, but that they are stochastically not smaller under the alternative. The problem of establishing the validity of randomized p‐values has received attention in previous literature. We show that our proposed randomized p‐values are valid under various discrete statistical models, which are such that the distribution of the corresponding test statistic belongs to an exponential family. The behavior of the power function for the tests based on the proposed randomized p‐values as a function of the sample size is also investigated. Simulations and a real data example are used to compare the different considered p‐values.</abstract><cop>Germany</cop><pub>Wiley - VCH Verlag GmbH &amp; Co. KGaA</pub><pmid>37857533</pmid><doi>10.1002/bimj.202300077</doi><tpages>20</tpages><orcidid>https://orcid.org/0000-0003-3084-3036</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0323-3847
ispartof Biometrical journal, 2024-01, Vol.66 (1), p.e2300077-n/a
issn 0323-3847
1521-4036
language eng
recordid cdi_proquest_miscellaneous_2880105322
source Wiley Online Library All Journals
subjects conservative tests
discretely distributed test statistics
group testing
Hypotheses
Mathematical models
multiple comparisons
Null hypothesis
Parameters
randomized tests
Statistical analysis
Statistical methods
Statistical models
Statistical tests
title Multiple testing of composite null hypotheses for discrete data using randomized p‐values
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-15T05%3A15%3A59IST&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=Multiple%20testing%20of%20composite%20null%20hypotheses%20for%20discrete%20data%20using%20randomized%20p%E2%80%90values&rft.jtitle=Biometrical%20journal&rft.au=Ochieng,%20Daniel&rft.date=2024-01&rft.volume=66&rft.issue=1&rft.spage=e2300077&rft.epage=n/a&rft.pages=e2300077-n/a&rft.issn=0323-3847&rft.eissn=1521-4036&rft_id=info:doi/10.1002/bimj.202300077&rft_dat=%3Cproquest_cross%3E2919357870%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=2919357870&rft_id=info:pmid/37857533&rfr_iscdi=true