A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model

A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model

In this paper, we discuss two-sample tests for high-dimension, non-Gaussian data. We suppose that two classes have a strongly spiked eigenvalue model. First, we investigate the noise space for high-dimension, non-Gaussian data. A two-sample test is proposed by using the cross-data-matrix (CDM) metho...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:JOURNAL OF THE JAPAN STATISTICAL SOCIETY 2017/12/28, Vol.47(2), pp.273-291
1. Verfasser: Ishii, Aki
Format: Artikel
Sprache:eng
Schlagworte:
Computer simulation
Cross-data-matrix methodology
eigenstructure
HDLSS
large p
small n
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 291
container_issue 2
container_start_page 273
container_title JOURNAL OF THE JAPAN STATISTICAL SOCIETY
container_volume 47
creator Ishii, Aki
description In this paper, we discuss two-sample tests for high-dimension, non-Gaussian data. We suppose that two classes have a strongly spiked eigenvalue model. First, we investigate the noise space for high-dimension, non-Gaussian data. A two-sample test is proposed by using the cross-data-matrix (CDM) methodology and its power is derived under some regularity conditions when the dimension is very large. We discuss the validity of assumptions. We check the performance of the proposed two-sample test procedure by simulations. Finally, we demonstrate the proposed two-sample test in actual data analyses.
doi_str_mv 10.14490/jjss.47.273
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2047266073</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2047266073</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3513-3625aa3e63fa3ecdfd10eef9acdf89ac9fe9ccb6b23d12bf1126374b17c80f1c3</originalsourceid><addsrcrecordid>eNpFkE9PwkAQxRujiYje_ACbeLW4_7otFxODCCaoB_C8Gbaz0Fq6uNtq-PZWMHh585L5zWTmRdE1owMm5ZDelWUIA5kOeCpOoh4TMouVUMlp57OMxzxN5Hl0EUJJqRCJkr0IHsi0WK3jx2KDdShcDRVZfLt4DptthWSBoSHWefLq6ngCbQgF1OQRGiBtnaMnQOaNd_Wq2pH5tvjAnIyLFdZfULVIXlyO1WV0ZqEKePVX-9H703gxmsazt8nz6GEWG5EwEQvFEwCBSthOTW5zRhHtEDqbdTq0ODRmqZZc5IwvLWNciVQuWWoyapkR_ejmsHfr3Wfb3a1L1_run6A5lSlXiqaio24PlPEuBI9Wb32xAb_TjOp9iPo3RC1Tzff4_QEvQwMrPMLgm8JU-M_-DRwbZg1eYy1-AJNYfOo</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2047266073</pqid></control><display><type>article</type><title>A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model</title><source>J-STAGE Free</source><source>EZB-FREE-00999 freely available EZB journals</source><creator>Ishii, Aki</creator><creatorcontrib>Ishii, Aki</creatorcontrib><description>In this paper, we discuss two-sample tests for high-dimension, non-Gaussian data. We suppose that two classes have a strongly spiked eigenvalue model. First, we investigate the noise space for high-dimension, non-Gaussian data. A two-sample test is proposed by using the cross-data-matrix (CDM) methodology and its power is derived under some regularity conditions when the dimension is very large. We discuss the validity of assumptions. We check the performance of the proposed two-sample test procedure by simulations. Finally, we demonstrate the proposed two-sample test in actual data analyses.</description><identifier>ISSN: 1882-2754</identifier><identifier>EISSN: 1348-6365</identifier><identifier>DOI: 10.14490/jjss.47.273</identifier><language>eng</language><publisher>Tokyo: THE JAPAN STATISTICAL SOCIETY</publisher><subject>Computer simulation ; Cross-data-matrix methodology ; eigenstructure ; HDLSS ; large p ; small n</subject><ispartof>JOURNAL OF THE JAPAN STATISTICAL SOCIETY, 2017/12/28, Vol.47(2), pp.273-291</ispartof><rights>2017 Japan Statistical Society</rights><rights>Copyright Japan Science and Technology Agency 2017</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c3513-3625aa3e63fa3ecdfd10eef9acdf89ac9fe9ccb6b23d12bf1126374b17c80f1c3</citedby><cites>FETCH-LOGICAL-c3513-3625aa3e63fa3ecdfd10eef9acdf89ac9fe9ccb6b23d12bf1126374b17c80f1c3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,776,780,1877,27901,27902</link.rule.ids></links><search><creatorcontrib>Ishii, Aki</creatorcontrib><title>A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model</title><title>JOURNAL OF THE JAPAN STATISTICAL SOCIETY</title><addtitle>JOURNAL OF THE JAPAN STATISTICAL SOCIETY</addtitle><description>In this paper, we discuss two-sample tests for high-dimension, non-Gaussian data. We suppose that two classes have a strongly spiked eigenvalue model. First, we investigate the noise space for high-dimension, non-Gaussian data. A two-sample test is proposed by using the cross-data-matrix (CDM) methodology and its power is derived under some regularity conditions when the dimension is very large. We discuss the validity of assumptions. We check the performance of the proposed two-sample test procedure by simulations. Finally, we demonstrate the proposed two-sample test in actual data analyses.</description><subject>Computer simulation</subject><subject>Cross-data-matrix methodology</subject><subject>eigenstructure</subject><subject>HDLSS</subject><subject>large p</subject><subject>small n</subject><issn>1882-2754</issn><issn>1348-6365</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><recordid>eNpFkE9PwkAQxRujiYje_ACbeLW4_7otFxODCCaoB_C8Gbaz0Fq6uNtq-PZWMHh585L5zWTmRdE1owMm5ZDelWUIA5kOeCpOoh4TMouVUMlp57OMxzxN5Hl0EUJJqRCJkr0IHsi0WK3jx2KDdShcDRVZfLt4DptthWSBoSHWefLq6ngCbQgF1OQRGiBtnaMnQOaNd_Wq2pH5tvjAnIyLFdZfULVIXlyO1WV0ZqEKePVX-9H703gxmsazt8nz6GEWG5EwEQvFEwCBSthOTW5zRhHtEDqbdTq0ODRmqZZc5IwvLWNciVQuWWoyapkR_ejmsHfr3Wfb3a1L1_run6A5lSlXiqaio24PlPEuBI9Wb32xAb_TjOp9iPo3RC1Tzff4_QEvQwMrPMLgm8JU-M_-DRwbZg1eYy1-AJNYfOo</recordid><startdate>20171228</startdate><enddate>20171228</enddate><creator>Ishii, Aki</creator><general>THE JAPAN STATISTICAL SOCIETY</general><general>Japan Science and Technology Agency</general><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20171228</creationdate><title>A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model</title><author>Ishii, Aki</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3513-3625aa3e63fa3ecdfd10eef9acdf89ac9fe9ccb6b23d12bf1126374b17c80f1c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Computer simulation</topic><topic>Cross-data-matrix methodology</topic><topic>eigenstructure</topic><topic>HDLSS</topic><topic>large p</topic><topic>small n</topic><toplevel>online_resources</toplevel><creatorcontrib>Ishii, Aki</creatorcontrib><collection>CrossRef</collection><jtitle>JOURNAL OF THE JAPAN STATISTICAL SOCIETY</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ishii, Aki</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model</atitle><jtitle>JOURNAL OF THE JAPAN STATISTICAL SOCIETY</jtitle><addtitle>JOURNAL OF THE JAPAN STATISTICAL SOCIETY</addtitle><date>2017-12-28</date><risdate>2017</risdate><volume>47</volume><issue>2</issue><spage>273</spage><epage>291</epage><pages>273-291</pages><issn>1882-2754</issn><eissn>1348-6365</eissn><abstract>In this paper, we discuss two-sample tests for high-dimension, non-Gaussian data. We suppose that two classes have a strongly spiked eigenvalue model. First, we investigate the noise space for high-dimension, non-Gaussian data. A two-sample test is proposed by using the cross-data-matrix (CDM) methodology and its power is derived under some regularity conditions when the dimension is very large. We discuss the validity of assumptions. We check the performance of the proposed two-sample test procedure by simulations. Finally, we demonstrate the proposed two-sample test in actual data analyses.</abstract><cop>Tokyo</cop><pub>THE JAPAN STATISTICAL SOCIETY</pub><doi>10.14490/jjss.47.273</doi><tpages>19</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1882-2754
ispartof JOURNAL OF THE JAPAN STATISTICAL SOCIETY, 2017/12/28, Vol.47(2), pp.273-291
issn 1882-2754
1348-6365
language eng
recordid cdi_proquest_journals_2047266073
source J-STAGE Free; EZB-FREE-00999 freely available EZB journals
subjects Computer simulation
Cross-data-matrix methodology
eigenstructure
HDLSS
large p
small n
title A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-06T12%3A51%3A38IST&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=A%20High-Dimensional%20Two-Sample%20Test%20for%20Non-Gaussian%20Data%20under%20a%20Strongly%20Spiked%20Eigenvalue%20Model&rft.jtitle=JOURNAL%20OF%20THE%20JAPAN%20STATISTICAL%20SOCIETY&rft.au=Ishii,%20Aki&rft.date=2017-12-28&rft.volume=47&rft.issue=2&rft.spage=273&rft.epage=291&rft.pages=273-291&rft.issn=1882-2754&rft.eissn=1348-6365&rft_id=info:doi/10.14490/jjss.47.273&rft_dat=%3Cproquest_cross%3E2047266073%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=2047266073&rft_id=info:pmid/&rfr_iscdi=true