Loading…
Random Effects Selection in Linear Mixed Models
We address the important practical problem of how to select the random effects component in a linear mixed model. A hierarchical Bayesian model is used to identify any random effect with zero variance. The proposed approach reparameterizes the mixed model so that functions of the covariance paramete...
Saved in:
Published in: | Biometrics 2003-12, Vol.59 (4), p.762-769 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143 |
---|---|
cites | cdi_FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143 |
container_end_page | 769 |
container_issue | 4 |
container_start_page | 762 |
container_title | Biometrics |
container_volume | 59 |
creator | Chen, Zhen Dunson, David B. |
description | We address the important practical problem of how to select the random effects component in a linear mixed model. A hierarchical Bayesian model is used to identify any random effect with zero variance. The proposed approach reparameterizes the mixed model so that functions of the covariance parameters of the random effects distribution are incorporated as regression coefficients on standard normal latent variables. We allow random effects to effectively drop out of the model by choosing mixture priors with point mass at zero for the random effects variances. Due to the reparameterization, the model enjoys a conditionally linear structure that facilitates the use of normal conjugate priors. We demonstrate that posterior computation can proceed via a simple and efficient Markov chain Monte Carlo algorithm. The methods are illustrated using simulated data and real data from a study relating prenatal exposure to polychlorinated biphenyls and psychomotor development of children. |
doi_str_mv | 10.1111/j.0006-341X.2003.00089.x |
format | article |
fullrecord | <record><control><sourceid>jstor_proqu</sourceid><recordid>TN_cdi_proquest_miscellaneous_71535744</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><jstor_id>3695314</jstor_id><sourcerecordid>3695314</sourcerecordid><originalsourceid>FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143</originalsourceid><addsrcrecordid>eNqNkEFPwjAUxxujEUS_gTE7eRu0a7vSgwdFRBKQBBUNl6ZbX5Ph2HCFCN_ejRG82svry_v9X9MfQh7BbVKezqKNMQ59yshnO8CYVm1XtrcnqEk4Iz5mAT5FzSPUQBfOLcpWchycowZhMpSM0ybqTHVm8qXXtxbitfNeIS1rkmdeknmjJANdeONkC8Yb5wZSd4nOrE4dXB1qC70_9d96z_5oMhj27kd-zCWXvokEyChgIQRxiG1IA8sEiQ02ENnYChpFkpiAaAkmBC0iG9GuCAzBtqslYbSFbuu9qyL_3oBbq2XiYkhTnUG-cUoQTrlgFditwbjInSvAqlWRLHWxUwSrSpZaqMqDqjyoSpbay1LbMnpzeGMTLcH8BQ92SuCuBn6SFHb_XqwehpNxeSvz13V-4dZ5cczTUHK6_6NfjxO3hu1xrIsvFQoquPp4Gaj5bDQl85lQj_QX_BiQYQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>71535744</pqid></control><display><type>article</type><title>Random Effects Selection in Linear Mixed Models</title><source>Wiley-Blackwell Journals</source><source>EBSCOhost SPORTDiscus with Full Text</source><source>JSTOR</source><creator>Chen, Zhen ; Dunson, David B.</creator><creatorcontrib>Chen, Zhen ; Dunson, David B.</creatorcontrib><description>We address the important practical problem of how to select the random effects component in a linear mixed model. A hierarchical Bayesian model is used to identify any random effect with zero variance. The proposed approach reparameterizes the mixed model so that functions of the covariance parameters of the random effects distribution are incorporated as regression coefficients on standard normal latent variables. We allow random effects to effectively drop out of the model by choosing mixture priors with point mass at zero for the random effects variances. Due to the reparameterization, the model enjoys a conditionally linear structure that facilitates the use of normal conjugate priors. We demonstrate that posterior computation can proceed via a simple and efficient Markov chain Monte Carlo algorithm. The methods are illustrated using simulated data and real data from a study relating prenatal exposure to polychlorinated biphenyls and psychomotor development of children.</description><identifier>ISSN: 0006-341X</identifier><identifier>EISSN: 1541-0420</identifier><identifier>DOI: 10.1111/j.0006-341X.2003.00089.x</identifier><identifier>PMID: 14969453</identifier><language>eng</language><publisher>350 Main Street , Malden , MA 02148 , U.S.A , and P.O. Box 1354, 9600 Garsington Road , Oxford OX4 2DQ , U.K: Blackwell Publishing</publisher><subject>Algorithms ; Bayes factor ; Biometrics ; Biometry ; Child ; Child Development - physiology ; Covariance ; Covariance matrices ; Epidemiology ; Homogeneity test ; Humans ; Latent variables ; Linear models ; Longitudinal data ; Markov Chains ; MCMC ; Model averaging ; Modeling ; Models, Statistical ; Monte Carlo Method ; Psychomotor development ; Regression coefficients ; Standard deviation ; Statistical variance ; Variable selection ; Variance components</subject><ispartof>Biometrics, 2003-12, Vol.59 (4), p.762-769</ispartof><rights>Copyright 2003 The International Biometric Society</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143</citedby><cites>FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1111%2Fj.0006-341X.2003.00089.x$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1111%2Fj.0006-341X.2003.00089.x$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>315,786,790,27957,27958,50923,51032,58593,58826</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/14969453$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Chen, Zhen</creatorcontrib><creatorcontrib>Dunson, David B.</creatorcontrib><title>Random Effects Selection in Linear Mixed Models</title><title>Biometrics</title><addtitle>Biometrics</addtitle><description>We address the important practical problem of how to select the random effects component in a linear mixed model. A hierarchical Bayesian model is used to identify any random effect with zero variance. The proposed approach reparameterizes the mixed model so that functions of the covariance parameters of the random effects distribution are incorporated as regression coefficients on standard normal latent variables. We allow random effects to effectively drop out of the model by choosing mixture priors with point mass at zero for the random effects variances. Due to the reparameterization, the model enjoys a conditionally linear structure that facilitates the use of normal conjugate priors. We demonstrate that posterior computation can proceed via a simple and efficient Markov chain Monte Carlo algorithm. The methods are illustrated using simulated data and real data from a study relating prenatal exposure to polychlorinated biphenyls and psychomotor development of children.</description><subject>Algorithms</subject><subject>Bayes factor</subject><subject>Biometrics</subject><subject>Biometry</subject><subject>Child</subject><subject>Child Development - physiology</subject><subject>Covariance</subject><subject>Covariance matrices</subject><subject>Epidemiology</subject><subject>Homogeneity test</subject><subject>Humans</subject><subject>Latent variables</subject><subject>Linear models</subject><subject>Longitudinal data</subject><subject>Markov Chains</subject><subject>MCMC</subject><subject>Model averaging</subject><subject>Modeling</subject><subject>Models, Statistical</subject><subject>Monte Carlo Method</subject><subject>Psychomotor development</subject><subject>Regression coefficients</subject><subject>Standard deviation</subject><subject>Statistical variance</subject><subject>Variable selection</subject><subject>Variance components</subject><issn>0006-341X</issn><issn>1541-0420</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2003</creationdate><recordtype>article</recordtype><recordid>eNqNkEFPwjAUxxujEUS_gTE7eRu0a7vSgwdFRBKQBBUNl6ZbX5Ph2HCFCN_ejRG82svry_v9X9MfQh7BbVKezqKNMQ59yshnO8CYVm1XtrcnqEk4Iz5mAT5FzSPUQBfOLcpWchycowZhMpSM0ybqTHVm8qXXtxbitfNeIS1rkmdeknmjJANdeONkC8Yb5wZSd4nOrE4dXB1qC70_9d96z_5oMhj27kd-zCWXvokEyChgIQRxiG1IA8sEiQ02ENnYChpFkpiAaAkmBC0iG9GuCAzBtqslYbSFbuu9qyL_3oBbq2XiYkhTnUG-cUoQTrlgFditwbjInSvAqlWRLHWxUwSrSpZaqMqDqjyoSpbay1LbMnpzeGMTLcH8BQ92SuCuBn6SFHb_XqwehpNxeSvz13V-4dZ5cczTUHK6_6NfjxO3hu1xrIsvFQoquPp4Gaj5bDQl85lQj_QX_BiQYQ</recordid><startdate>200312</startdate><enddate>200312</enddate><creator>Chen, Zhen</creator><creator>Dunson, David B.</creator><general>Blackwell Publishing</general><general>International Biometric Society</general><scope>BSCLL</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope></search><sort><creationdate>200312</creationdate><title>Random Effects Selection in Linear Mixed Models</title><author>Chen, Zhen ; Dunson, David B.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2003</creationdate><topic>Algorithms</topic><topic>Bayes factor</topic><topic>Biometrics</topic><topic>Biometry</topic><topic>Child</topic><topic>Child Development - physiology</topic><topic>Covariance</topic><topic>Covariance matrices</topic><topic>Epidemiology</topic><topic>Homogeneity test</topic><topic>Humans</topic><topic>Latent variables</topic><topic>Linear models</topic><topic>Longitudinal data</topic><topic>Markov Chains</topic><topic>MCMC</topic><topic>Model averaging</topic><topic>Modeling</topic><topic>Models, Statistical</topic><topic>Monte Carlo Method</topic><topic>Psychomotor development</topic><topic>Regression coefficients</topic><topic>Standard deviation</topic><topic>Statistical variance</topic><topic>Variable selection</topic><topic>Variance components</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Chen, Zhen</creatorcontrib><creatorcontrib>Dunson, David B.</creatorcontrib><collection>Istex</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Biometrics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Chen, Zhen</au><au>Dunson, David B.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Random Effects Selection in Linear Mixed Models</atitle><jtitle>Biometrics</jtitle><addtitle>Biometrics</addtitle><date>2003-12</date><risdate>2003</risdate><volume>59</volume><issue>4</issue><spage>762</spage><epage>769</epage><pages>762-769</pages><issn>0006-341X</issn><eissn>1541-0420</eissn><notes>istex:2D768C496538415C466C6461963DF082509B655C</notes><notes>ArticleID:BIOM089</notes><notes>ark:/67375/WNG-ZVLR1ZV7-D</notes><notes>ObjectType-Article-1</notes><notes>SourceType-Scholarly Journals-1</notes><notes>ObjectType-Feature-2</notes><notes>content type line 23</notes><abstract>We address the important practical problem of how to select the random effects component in a linear mixed model. A hierarchical Bayesian model is used to identify any random effect with zero variance. The proposed approach reparameterizes the mixed model so that functions of the covariance parameters of the random effects distribution are incorporated as regression coefficients on standard normal latent variables. We allow random effects to effectively drop out of the model by choosing mixture priors with point mass at zero for the random effects variances. Due to the reparameterization, the model enjoys a conditionally linear structure that facilitates the use of normal conjugate priors. We demonstrate that posterior computation can proceed via a simple and efficient Markov chain Monte Carlo algorithm. The methods are illustrated using simulated data and real data from a study relating prenatal exposure to polychlorinated biphenyls and psychomotor development of children.</abstract><cop>350 Main Street , Malden , MA 02148 , U.S.A , and P.O. Box 1354, 9600 Garsington Road , Oxford OX4 2DQ , U.K</cop><pub>Blackwell Publishing</pub><pmid>14969453</pmid><doi>10.1111/j.0006-341X.2003.00089.x</doi><tpages>8</tpages><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0006-341X |
ispartof | Biometrics, 2003-12, Vol.59 (4), p.762-769 |
issn | 0006-341X 1541-0420 |
language | eng |
recordid | cdi_proquest_miscellaneous_71535744 |
source | Wiley-Blackwell Journals; EBSCOhost SPORTDiscus with Full Text; JSTOR |
subjects | Algorithms Bayes factor Biometrics Biometry Child Child Development - physiology Covariance Covariance matrices Epidemiology Homogeneity test Humans Latent variables Linear models Longitudinal data Markov Chains MCMC Model averaging Modeling Models, Statistical Monte Carlo Method Psychomotor development Regression coefficients Standard deviation Statistical variance Variable selection Variance components |
title | Random Effects Selection in Linear Mixed Models |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-09-22T20%3A24%3A39IST&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=Random%20Effects%20Selection%20in%20Linear%20Mixed%20Models&rft.jtitle=Biometrics&rft.au=Chen,%20Zhen&rft.date=2003-12&rft.volume=59&rft.issue=4&rft.spage=762&rft.epage=769&rft.pages=762-769&rft.issn=0006-341X&rft.eissn=1541-0420&rft_id=info:doi/10.1111/j.0006-341X.2003.00089.x&rft_dat=%3Cjstor_proqu%3E3695314%3C/jstor_proqu%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c5959-db7e9b246e2c60f632f471cd0debfcf73bb91d21a9ed6ea7bfb3872d10f8a9143%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=71535744&rft_id=info:pmid/14969453&rft_jstor_id=3695314&rfr_iscdi=true |