Alternatives to the usual likelihood ratio test in mixed linear models
The small-sample performance of alternatives to the usual likelihood ratio test in mixed linear models is investigated. Specifically, the following tests for fixed effects are considered: (i) a bootstrap-based test, (ii) the Bartlett-corrected usual test, and (iii) an adjusted profile likelihood rat...
Gespeichert in:
| Veröffentlicht in: | Computational statistics & data analysis 2014-01, Vol.69, p.184-197 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 197 |
|---|---|
| container_issue | |
| container_start_page | 184 |
| container_title | Computational statistics & data analysis |
| container_volume | 69 |
| creator | Stein, Markus Chagas da Silva, Michel Ferreira Duczmal, Luiz Henrique |
| description | The small-sample performance of alternatives to the usual likelihood ratio test in mixed linear models is investigated. Specifically, the following tests for fixed effects are considered: (i) a bootstrap-based test, (ii) the Bartlett-corrected usual test, and (iii) an adjusted profile likelihood ratio test. The last test is derived using an approximation to the modified profile likelihood proposed by Barndorff-Nielsen, based on the work of Severini. Bootstrap resampling is performed to numerically construct a Bartlett correction factor for the usual test statistic, and also to obtain a critical value that does not rely on first-order asymptotics. The numerical evidence presented in the paper slightly favors the Bartlett-corrected usual test. An application to real longitudinal data is presented. |
| doi_str_mv | 10.1016/j.csda.2013.08.002 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_1506381097</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0167947313002880</els_id><sourcerecordid>1506381097</sourcerecordid><originalsourceid>FETCH-LOGICAL-c399t-e4f2853f19dbcb87378ab7cc3afbead2f2f90a1c3aa28de22e12a2e1e10c57ed3</originalsourceid><addsrcrecordid>eNp9kE1LAzEQhoMoWKt_wFOOXnbNR7fJgpdS_IKCFz2HbDJLU7ObmmSL_ntT6tnLDMP7vMPMi9AtJTUldHm_q02yumaE8prImhB2hmZUClYJ3rBzNCuQqNqF4JfoKqUdKcRCyBl6WvkMcdTZHSDhHHDeAp7SpD327hO824ZgcSx6kSBl7EY8uG-wRR5BRzwECz5do4te-wQ3f32OPp4e39cv1ebt-XW92lSGt22uYNEz2fCetrYznRRcSN0JY7juO9CW9axviaZl1kxaYAwo06UAJaYRYPkc3Z327mP4mso9anDJgPd6hDAlRRuy5JKSVhSUnVATQ0oRerWPbtDxR1GijqGpnTqGpo6hKSJViaSYHk6m8hMcHESVjIPRgHURTFY2uP_svyatdtY</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1506381097</pqid></control><display><type>article</type><title>Alternatives to the usual likelihood ratio test in mixed linear models</title><source>Elsevier ScienceDirect Journals</source><creator>Stein, Markus Chagas ; da Silva, Michel Ferreira ; Duczmal, Luiz Henrique</creator><creatorcontrib>Stein, Markus Chagas ; da Silva, Michel Ferreira ; Duczmal, Luiz Henrique</creatorcontrib><description>The small-sample performance of alternatives to the usual likelihood ratio test in mixed linear models is investigated. Specifically, the following tests for fixed effects are considered: (i) a bootstrap-based test, (ii) the Bartlett-corrected usual test, and (iii) an adjusted profile likelihood ratio test. The last test is derived using an approximation to the modified profile likelihood proposed by Barndorff-Nielsen, based on the work of Severini. Bootstrap resampling is performed to numerically construct a Bartlett correction factor for the usual test statistic, and also to obtain a critical value that does not rely on first-order asymptotics. The numerical evidence presented in the paper slightly favors the Bartlett-corrected usual test. An application to real longitudinal data is presented.</description><identifier>ISSN: 0167-9473</identifier><identifier>EISSN: 1872-7352</identifier><identifier>DOI: 10.1016/j.csda.2013.08.002</identifier><language>eng</language><publisher>Elsevier B.V</publisher><subject>Adjusted profile likelihood ; Adjustment ; Bartlett correction ; Bootstrap ; Likelihood ratio test ; Mixed linear model</subject><ispartof>Computational statistics & data analysis, 2014-01, Vol.69, p.184-197</ispartof><rights>2013 Elsevier B.V.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c399t-e4f2853f19dbcb87378ab7cc3afbead2f2f90a1c3aa28de22e12a2e1e10c57ed3</citedby><cites>FETCH-LOGICAL-c399t-e4f2853f19dbcb87378ab7cc3afbead2f2f90a1c3aa28de22e12a2e1e10c57ed3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.sciencedirect.com/science/article/pii/S0167947313002880$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,776,780,3537,27901,27902,65306</link.rule.ids></links><search><creatorcontrib>Stein, Markus Chagas</creatorcontrib><creatorcontrib>da Silva, Michel Ferreira</creatorcontrib><creatorcontrib>Duczmal, Luiz Henrique</creatorcontrib><title>Alternatives to the usual likelihood ratio test in mixed linear models</title><title>Computational statistics & data analysis</title><description>The small-sample performance of alternatives to the usual likelihood ratio test in mixed linear models is investigated. Specifically, the following tests for fixed effects are considered: (i) a bootstrap-based test, (ii) the Bartlett-corrected usual test, and (iii) an adjusted profile likelihood ratio test. The last test is derived using an approximation to the modified profile likelihood proposed by Barndorff-Nielsen, based on the work of Severini. Bootstrap resampling is performed to numerically construct a Bartlett correction factor for the usual test statistic, and also to obtain a critical value that does not rely on first-order asymptotics. The numerical evidence presented in the paper slightly favors the Bartlett-corrected usual test. An application to real longitudinal data is presented.</description><subject>Adjusted profile likelihood</subject><subject>Adjustment</subject><subject>Bartlett correction</subject><subject>Bootstrap</subject><subject>Likelihood ratio test</subject><subject>Mixed linear model</subject><issn>0167-9473</issn><issn>1872-7352</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2014</creationdate><recordtype>article</recordtype><recordid>eNp9kE1LAzEQhoMoWKt_wFOOXnbNR7fJgpdS_IKCFz2HbDJLU7ObmmSL_ntT6tnLDMP7vMPMi9AtJTUldHm_q02yumaE8prImhB2hmZUClYJ3rBzNCuQqNqF4JfoKqUdKcRCyBl6WvkMcdTZHSDhHHDeAp7SpD327hO824ZgcSx6kSBl7EY8uG-wRR5BRzwECz5do4te-wQ3f32OPp4e39cv1ebt-XW92lSGt22uYNEz2fCetrYznRRcSN0JY7juO9CW9axviaZl1kxaYAwo06UAJaYRYPkc3Z327mP4mso9anDJgPd6hDAlRRuy5JKSVhSUnVATQ0oRerWPbtDxR1GijqGpnTqGpo6hKSJViaSYHk6m8hMcHESVjIPRgHURTFY2uP_svyatdtY</recordid><startdate>201401</startdate><enddate>201401</enddate><creator>Stein, Markus Chagas</creator><creator>da Silva, Michel Ferreira</creator><creator>Duczmal, Luiz Henrique</creator><general>Elsevier B.V</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>201401</creationdate><title>Alternatives to the usual likelihood ratio test in mixed linear models</title><author>Stein, Markus Chagas ; da Silva, Michel Ferreira ; Duczmal, Luiz Henrique</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c399t-e4f2853f19dbcb87378ab7cc3afbead2f2f90a1c3aa28de22e12a2e1e10c57ed3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2014</creationdate><topic>Adjusted profile likelihood</topic><topic>Adjustment</topic><topic>Bartlett correction</topic><topic>Bootstrap</topic><topic>Likelihood ratio test</topic><topic>Mixed linear model</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Stein, Markus Chagas</creatorcontrib><creatorcontrib>da Silva, Michel Ferreira</creatorcontrib><creatorcontrib>Duczmal, Luiz Henrique</creatorcontrib><collection>CrossRef</collection><collection>Computer and Information Systems 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>Computational statistics & data analysis</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Stein, Markus Chagas</au><au>da Silva, Michel Ferreira</au><au>Duczmal, Luiz Henrique</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Alternatives to the usual likelihood ratio test in mixed linear models</atitle><jtitle>Computational statistics & data analysis</jtitle><date>2014-01</date><risdate>2014</risdate><volume>69</volume><spage>184</spage><epage>197</epage><pages>184-197</pages><issn>0167-9473</issn><eissn>1872-7352</eissn><abstract>The small-sample performance of alternatives to the usual likelihood ratio test in mixed linear models is investigated. Specifically, the following tests for fixed effects are considered: (i) a bootstrap-based test, (ii) the Bartlett-corrected usual test, and (iii) an adjusted profile likelihood ratio test. The last test is derived using an approximation to the modified profile likelihood proposed by Barndorff-Nielsen, based on the work of Severini. Bootstrap resampling is performed to numerically construct a Bartlett correction factor for the usual test statistic, and also to obtain a critical value that does not rely on first-order asymptotics. The numerical evidence presented in the paper slightly favors the Bartlett-corrected usual test. An application to real longitudinal data is presented.</abstract><pub>Elsevier B.V</pub><doi>10.1016/j.csda.2013.08.002</doi><tpages>14</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0167-9473 |
| ispartof | Computational statistics & data analysis, 2014-01, Vol.69, p.184-197 |
| issn | 0167-9473 1872-7352 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_1506381097 |
| source | Elsevier ScienceDirect Journals |
| subjects | Adjusted profile likelihood Adjustment Bartlett correction Bootstrap Likelihood ratio test Mixed linear model |
| title | Alternatives to the usual likelihood ratio test in mixed linear models |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-08T20%3A24%3A16IST&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=Alternatives%20to%20the%20usual%20likelihood%20ratio%20test%20in%20mixed%20linear%20models&rft.jtitle=Computational%20statistics%20&%20data%20analysis&rft.au=Stein,%20Markus%20Chagas&rft.date=2014-01&rft.volume=69&rft.spage=184&rft.epage=197&rft.pages=184-197&rft.issn=0167-9473&rft.eissn=1872-7352&rft_id=info:doi/10.1016/j.csda.2013.08.002&rft_dat=%3Cproquest_cross%3E1506381097%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=1506381097&rft_id=info:pmid/&rft_els_id=S0167947313002880&rfr_iscdi=true |