A Parallel Algorithm for Large-Scale Nonconvex Penalized Quantile Regression

Penalized quantile regression (PQR) provides a useful tool for analyzing high-dimensional data with heterogeneity. However, its computation is challenging due to the nonsmoothness and (sometimes) the nonconvexity of the objective function. An iterative coordinate descent algorithm (QICD) was recentl...

Full description

Saved in:

Bibliographic Details
Published in:	Journal of computational and graphical statistics Vol. 26; no. 4; pp. 935 - 939
Main Authors:	Yu, Liqun, Lin, Nan, Wang, Lan
Format:	Journal Article
Language:	English
Published:	Alexandria Taylor & Francis 02.10.2017 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America Taylor & Francis Ltd
Subjects:	ADMM Algorithms Computation Descent Dimensional analysis Iterative methods Nonconvex penalty Parallelization Quantile regression and single-loop algorithm Regression analysis Short Technical Notes Statistical analysis Studies
ISSN:	1061-8600, 1537-2715
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Penalized quantile regression (PQR) provides a useful tool for analyzing high-dimensional data with heterogeneity. However, its computation is challenging due to the nonsmoothness and (sometimes) the nonconvexity of the objective function. An iterative coordinate descent algorithm (QICD) was recently proposed to solve PQR with nonconvex penalty. The QICD significantly improves the computational speed but requires a double-loop. In this article, we propose an alternative algorithm based on the alternating direction method of multiplier (ADMM). By writing the PQR into a special ADMM form, we can solve the iterations exactly without using coordinate descent. This results in a new single-loop algorithm, which we refer to as the QPADM algorithm. The QPADM demonstrates favorable performance in both computational speed and statistical accuracy, particularly when the sample size n and/or the number of features p are large. Supplementary material for this article is available online.
AbstractList	Penalized quantile regression (PQR) provides a useful tool for analyzing high-dimensional data with heterogeneity. However, its computation is challenging due to the nonsmoothness and (sometimes) the nonconvexity of the objective function. An iterative coordinate descent algorithm (QICD) was recently proposed to solve PQR with nonconvex penalty. The QICD significantly improves the computational speed but requires a double-loop. In this article, we propose an alternative algorithm based on the alternating direction method of multiplier (ADMM). By writing the PQR into a special ADMM form, we can solve the iterations exactly without using coordinate descent. This results in a new single-loop algorithm, which we refer to as the QPADM algorithm. The QPADM demonstrates favorable performance in both computational speed and statistical accuracy, particularly when the sample size n and/or the number of features p are large. Penalized quantile regression (PQR) provides a useful tool for analyzing high-dimensional data with heterogeneity. However, its computation is challenging due to the nonsmoothness and (sometimes) the nonconvexity of the objective function. An iterative coordinate descent algorithm (QICD) was recently proposed to solve PQR with nonconvex penalty. The QICD significantly improves the computational speed but requires a double-loop. In this article, we propose an alternative algorithm based on the alternating direction method of multiplier (ADMM). By writing the PQR into a special ADMM form, we can solve the iterations exactly without using coordinate descent. This results in a new single-loop algorithm, which we refer to as the QPADM algorithm. The QPADM demonstrates favorable performance in both computational speed and statistical accuracy, particularly when the sample size n and/or the number of features p are large. Supplementary material for this article is available online.
Author	Yu, Liqun Wang, Lan Lin, Nan
Author_xml	– sequence: 1 givenname: Liqun surname: Yu fullname: Yu, Liqun organization: Department of Mathematics, Washington University in St. Louis – sequence: 2 givenname: Nan surname: Lin fullname: Lin, Nan email: nlin@wustl.edu organization: Department of Mathematics, Washington University in St. Louis – sequence: 3 givenname: Lan surname: Wang fullname: Wang, Lan organization: School of Statistics, University of Minnesota
BookMark	eNqFkFtLAzEQhYMoeP0JwoLPW3PZZjf4YhFvULz7HKbpbE1JE01Stf56t1R98EGfZmDOmTnzbZN1HzwSss9oj9GGHjIqWSMp7XHK6h4TvBFSrpEt1hd1yWvWX-_6TlMuRZtkO6UppZRJVW-R4aC4gQjOoSsGbhKizU-zog2xGEKcYHlvwGFxFbwJ_hXfixv04OwHjovbOfhsu-EdTiKmZIPfJRstuIR7X3WHPJ6dPpxclMPr88uTwbA0Fa1zCQJR0pYbgaqi_Ro4wwroSODYcIFGjCSrWGVAcWhHXLUVoAKFqmVcNXIkdsjBau9zDC9zTFlPwzx2wZJmSnEpVMN4p-qvVCaGlCK2-jnaGcSFZlQvwelvcHoJTn-B63xHv3zGZsjdfzmCdf-691fuacoh_pysqkZyypepjldz6zvKM3gL0Y11hoULsY3gjU1a_H3iE__bkdo
CitedBy_id	crossref_primary_10_1016_j_jspi_2024_106144 crossref_primary_10_1080_00949655_2018_1448397 crossref_primary_10_1016_j_jspi_2024_106186 crossref_primary_10_1080_02331888_2025_2482070 crossref_primary_10_1007_s00184_019_00744_3 crossref_primary_10_3390_math12050735 crossref_primary_10_1007_s10618_022_00914_4 crossref_primary_10_1007_s11222_025_10631_9 crossref_primary_10_1016_j_csda_2023_107812 crossref_primary_10_1016_j_cam_2023_115192 crossref_primary_10_1080_10618600_2024_2308798 crossref_primary_10_1177_09622802211060520 crossref_primary_10_1214_22_AOS2214 crossref_primary_10_1080_03610918_2021_1888998 crossref_primary_10_1109_TSP_2023_3325622 crossref_primary_10_1016_j_jeconom_2024_105791 crossref_primary_10_1016_j_jeconom_2023_01_028 crossref_primary_10_1016_j_ins_2023_119259 crossref_primary_10_1007_s00362_021_01229_0 crossref_primary_10_1080_03610918_2023_2249271 crossref_primary_10_1016_j_csda_2024_108081 crossref_primary_10_1016_j_jspi_2021_07_003 crossref_primary_10_1002_sam_11700 crossref_primary_10_1080_10618600_2023_2275999 crossref_primary_10_3390_en15114135 crossref_primary_10_1080_1331677X_2021_1978306 crossref_primary_10_1080_10618600_2020_1840996 crossref_primary_10_1016_j_csda_2023_107901 crossref_primary_10_1214_18_AOS1777 crossref_primary_10_1155_2021_6341707 crossref_primary_10_1016_j_csda_2019_106819 crossref_primary_10_1007_s00180_022_01318_0 crossref_primary_10_1007_s11424_024_3510_8 crossref_primary_10_1016_j_neucom_2024_128671 crossref_primary_10_1007_s10114_025_3390_4 crossref_primary_10_1016_j_csda_2021_107388 crossref_primary_10_1080_07474938_2025_2503355
Cites_doi	10.1016/0898-1221(76)90003-1 10.1198/016214501753382273 10.1137/140990309 10.1137/140998135 10.1145/1327452.1327492 10.2307/1913643 10.1080/01621459.2012.656014 10.1111/j.2517-6161.1996.tb02080.x 10.1214/10-AOS827 10.1214/09-AOS729 10.1051/m2an/197509R200411 10.1080/10618600.2014.913516 10.1080/01621459.2013.836975 10.1561/2200000016
ContentType	Journal Article
Copyright	2017 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America 2017 2017 American Statistical Association, the Institute of Mathematical Statistics, and the Interface Foundation of North America Copyright American Statistical Association 2017
Copyright_xml	– notice: 2017 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America 2017 – notice: 2017 American Statistical Association, the Institute of Mathematical Statistics, and the Interface Foundation of North America – notice: Copyright American Statistical Association 2017
DBID	AAYXX CITATION JQ2
DOI	10.1080/10618600.2017.1328366
DatabaseName	CrossRef ProQuest Computer Science Collection
DatabaseTitle	CrossRef ProQuest Computer Science Collection
DatabaseTitleList	ProQuest Computer Science Collection
DeliveryMethod	fulltext_linktorsrc
Discipline	Statistics Mathematics
EISSN	1537-2715
EndPage	939
ExternalDocumentID	10_1080_10618600_2017_1328366 44862022 1328366
Genre	Note
GroupedDBID	-~X .4S .7F .DC .QJ 0BK 0R~ 2AX 30N 4.4 5GY AAENE AAGDL AAHIA AAJMT AALDU AAMIU AAPUL AAQRR AAWIL ABAWQ ABBHK ABCCY ABFAN ABFIM ABJNI ABLIJ ABLJU ABPAQ ABPEM ABQDR ABTAI ABXSQ ABXUL ABXYU ABYWD ACDIW ACGFO ACGFS ACHJO ACIWK ACMTB ACTIO ACTMH ADCVX ADGTB ADODI ADXHL AEGXH AELLO AENEX AEOZL AEPSL AEUPB AEYOC AFRVT AFVYC AGDLA AGLNM AGMYJ AHDZW AIAGR AIHAF AIJEM AKBRZ AKBVH AKOOK ALMA_UNASSIGNED_HOLDINGS ALQZU ALRMG AMVHM AQRUH AQTUD ARCSS AVBZW AWYRJ BLEHA CCCUG CS3 D0L DGEBU DKSSO DQDLB DSRWC DU5 EBS ECEWR EJD E~A E~B F5P GTTXZ H13 HF~ HQ6 HZ~ H~P IPNFZ IPSME J.P JAA JAAYA JBMMH JBZCM JENOY JHFFW JKQEH JLEZI JLXEF JMS JPL JST KYCEM LJTGL M4Z MS~ NA5 NY~ O9- P2P PQQKQ RIG RNANH ROSJB RTWRZ RWL RXW S-T SA0 SNACF TAE TASJS TBQAZ TDBHL TEJ TFL TFT TFW TN5 TTHFI TUROJ TUS UT5 UU3 WZA XWC ZGOLN ~S~ ADYSH AMPGV AAYXX CITATION JQ2
ID	FETCH-LOGICAL-c407t-a3ee60f2c3e94057a21e4a0b3edc23ec3b61414ca92afb29f4ae9a9e9f12986b3
IEDL.DBID	TFW
ISICitedReferencesCount	46
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000423019700023&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1061-8600
IngestDate	Mon Nov 10 03:09:36 EST 2025 Tue Nov 18 21:09:32 EST 2025 Sat Nov 29 03:24:16 EST 2025 Thu May 29 09:14:43 EDT 2025 Mon Oct 20 23:49:31 EDT 2025
IsDoiOpenAccess	false
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	4
Language	English
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c407t-a3ee60f2c3e94057a21e4a0b3edc23ec3b61414ca92afb29f4ae9a9e9f12986b3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
OpenAccessLink	https://figshare.com/articles/journal_contribution/A_Parallel_Algorithm_for_Large-scale_Nonconvex_Penalized_Quantile_Regression/5024429
PQID	1992639812
PQPubID	29738
PageCount	5
ParticipantIDs	informaworld_taylorfrancis_310_1080_10618600_2017_1328366 proquest_journals_1992639812 crossref_primary_10_1080_10618600_2017_1328366 crossref_citationtrail_10_1080_10618600_2017_1328366 jstor_primary_44862022
PublicationCentury	2000
PublicationDate	2017-10-02
PublicationDateYYYYMMDD	2017-10-02
PublicationDate_xml	– month: 10 year: 2017 text: 2017-10-02 day: 02
PublicationDecade	2010
PublicationPlace	Alexandria
PublicationPlace_xml	– name: Alexandria
PublicationTitle	Journal of computational and graphical statistics
PublicationYear	2017
Publisher	Taylor & Francis American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America Taylor & Francis Ltd
Publisher_xml	– name: Taylor & Francis – name: American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America – name: Taylor & Francis Ltd
References	cit0011 cit0001 cit0023 cit0010 Koenker R. (cit0008) 2008 Sherwood B. (cit0016) 2016 Woodbury M. A. (cit0021) 1950; 42 Wang Y. (cit0019) 2015 Mota J. F. (cit0012) 2011 Zaharia M. (cit0022) 2010 Tibshirani R. (cit0017) 1996; 58 (cit0015) 2008 cit0009 cit0006 cit0007 cit0018 cit0004 cit0005 cit0002 cit0013 White T. (cit0020) 2012 cit0003 cit0014
References_xml	– ident: cit0013 – ident: cit0005 doi: 10.1016/0898-1221(76)90003-1 – year: 2008 ident: cit0008 publication-title: http://CRAN.R-project.org/package=quantre – ident: cit0004 doi: 10.1198/016214501753382273 – ident: cit0007 doi: 10.1137/140990309 – year: 2016 ident: cit0016 publication-title: R Package Version 1.4, R Foundation for Statistical Computing, Vienna, Austria – ident: cit0011 doi: 10.1137/140998135 – volume: 42 start-page: 106 year: 1950 ident: cit0021 publication-title: Memorandum Report – ident: cit0003 doi: 10.1145/1327452.1327492 – ident: cit0009 doi: 10.2307/1913643 – ident: cit0018 doi: 10.1080/01621459.2012.656014 – volume-title: Hadoop: The Definitive Guide year: 2012 ident: cit0020 – volume: 58 start-page: 267 year: 1996 ident: cit0017 publication-title: Journal of the Royal Statistical Society doi: 10.1111/j.2517-6161.1996.tb02080.x – volume-title: R: A Language and Environment for Statistical Computing year: 2008 ident: cit0015 – ident: cit0001 doi: 10.1214/10-AOS827 – ident: cit0023 doi: 10.1214/09-AOS729 – ident: cit0006 doi: 10.1051/m2an/197509R200411 – year: 2011 ident: cit0012 publication-title: arXiv preprint arXiv:1112.2295 – ident: cit0014 doi: 10.1080/10618600.2014.913516 – ident: cit0010 doi: 10.1080/01621459.2013.836975 – year: 2015 ident: cit0019 publication-title: arXiv preprint arXiv:1511.06324 – ident: cit0002 doi: 10.1561/2200000016 – start-page: 10 year: 2010 ident: cit0022 publication-title: Proceedings of the 2nd USENIX Conference on Hot Topics in Cloud Computing
SSID	ssj0001697
Score	2.413928
Snippet	Penalized quantile regression (PQR) provides a useful tool for analyzing high-dimensional data with heterogeneity. However, its computation is challenging due...
SourceID	proquest crossref jstor informaworld
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	935
SubjectTerms	ADMM Algorithms Computation Descent Dimensional analysis Iterative methods Nonconvex penalty Parallelization Quantile regression and single-loop algorithm Regression analysis Short Technical Notes Statistical analysis Studies
Title	A Parallel Algorithm for Large-Scale Nonconvex Penalized Quantile Regression
URI	https://www.tandfonline.com/doi/abs/10.1080/10618600.2017.1328366 https://www.jstor.org/stable/44862022 https://www.proquest.com/docview/1992639812
Volume	26
WOSCitedRecordID	wos000423019700023&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAWR databaseName: Taylor & Francis Online Journals customDbUrl: eissn: 1537-2715 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001697 issn: 1061-8600 databaseCode: TFW dateStart: 19920301 isFulltext: true titleUrlDefault: https://www.tandfonline.com providerName: Taylor & Francis
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV09T8MwELVQxVAGPgoVhYI8sLo0dkjisUJUDKUqUKBbZLtOQSop6gdC_HruEqeiQqgDjBnOTuzz3bvk5R0hZ5DDdciNz1TIA-brJGS6KTSLPAPJRxnpZy_cHjthtxsNBrLn2IQzR6vEGjrJhSKyWI2HW-lZwYg7xyomgkSNxKywAeVUJAIU3YbUj0ez335axmLPtVcBC4YmxT88v42ykp1WtEsLvuKPmJ0lovbOPzzCLtl2KJS2crfZIxs2rZCtm6WE66xCyghDcxXnfdJp0Z6aYtsVMBqPJtOX-fMrhdumHWSSs3vYaUu7kzQjsX_QnkV8_2mH9HYBWweRh97ZUU65TQ_IQ_uqf3nNXB8GZqDcmzMlrA2aCTfCSsR3invWV00t7NBwYY3QkOM93yjJVaK5THxlpZJWJgAmokCLKimlk9QeEqpDwEeBwC4zMJSvowD7fiQXIVf4oUHUiF-sf2ycSDn2yhjHntMyLVYuxpWL3crVSGNp9pardKwzkN83N55nr0eSvJdJLNbYVjNPWM4EFW7AAQvVSL1wjdhFglmM9F5AgYCjjv4w5zEp42XGIuR1UppPF_aEbJp3cITpaebzX0lL-Cg
linkProvider	Taylor & Francis
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3JTsQwDLXYJODAjhjWHLgGpklpmyNCjECUEcuw3KIkpIAEHTQMCPH12F1GIIQ4wLly0saO_Zy-2ACbGMNtLFzITSwiHtos5rYpLU8Ch8HHOBUWB26XadxuJ9fX6vNdGKJVUg6dlYUiCl9Nm5sOo2tK3DalMQlGamJmxVuYTyUyioZhdAdjLdXP77SuBt44qBqsoAgnmfoWz0_DfIlPX6qX1ozFb167CEWt6f_4iBmYqoAo2y0tZxaGfD4Hk8eDKq7PczBBSLQs5DwP6S47MT3qvIJCD7fd3n3_7pHhe7OUyOT8HJXtWbubFzz2N3biCeK_-xt2-oLaQ-fDzvxtybrNF-Citd_ZO-BVKwbuMOPrcyO9j5qZcNIrgnhGBD40TSv9jRPSO2kxzAehM0qYzAqVhcYro7zKEE8kkZWLMJJ3c78EzMYIkSJJjWZwqNAmEbX-yHZiYehfg2xAWCtAu6pOObXLeNBBVc60XjlNK6erlWvA1kDsqSzU8ZuA-qxd3S9OSLKynYmWv8guFqYwmAmT3EggHGrAam0bunIGz5oYvggEEUot_2HODRg_6BynOj1sH63ABD0qSIViFUb6vRe_BmPuFY2it15sgA8WZPxS
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LT9wwEB5RqBA9lLe6LQ8fuBo2durER9R21YpltbSUcrNsrw2VaBbtLlXVX9-ZxFkVoYoDPUdjJzOTmW-cLzMAB5jDXSF8zm0hFM9dLLjrSsfLzGPysV7n9YHbRb8YDMrLSz1MbMJpolVSDR2bRhF1rKaX-3YUW0bcEVUxJSZqImYVh1hOlVKpZ7CE0FmRk5_3vs2DcZbmq6AIJ5n2J55_LXMvPd1rXtoSFh8E7ToT9Vb_wzOswcsEQ9lx4zfrsBCqDXhxOu_hOt2AFcKhTRvnTegfs6Gd0NwVFLq5Gk--z65_MLxt1icqOf-Cpg5sMK5qFvsvNgwE8H-HETu7Q9th6GGfw1XDua224Gvvw_m7jzwNYuAe670ZtzIE1Y3Cy6AJ4FmRhdx2nQwjL2Tw0mGSz3JvtbDRCR1zG7TVQUdEE6VychsWq3EVXgFzBQIkJWnMDC6Vu1LR4I_4thCWvjTIDuSt_o1PXcppWMaNyVIz01ZzhjRnkuY6cDgXu23adDwmoP82rpnV5yOxGWZi5COy27UnzHfCElcJBEMd2Gldw6RQMDXE70UYiEDq9RP23Ifl4fue6X8anLyBFbpSMwrFDizOJndhF577n-gTk73a_f8ADw_7BA
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+Parallel+Algorithm+for+Large-Scale+Nonconvex+Penalized+Quantile+Regression&rft.jtitle=Journal+of+computational+and+graphical+statistics&rft.au=Yu%2C+Liqun&rft.au=Lin%2C+Nan&rft.au=Wang%2C+Lan&rft.date=2017-10-02&rft.pub=Taylor+%26+Francis&rft.issn=1061-8600&rft.eissn=1537-2715&rft.volume=26&rft.issue=4&rft.spage=935&rft.epage=939&rft_id=info:doi/10.1080%2F10618600.2017.1328366&rft.externalDocID=1328366
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1061-8600&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1061-8600&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1061-8600&client=summon