{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T13:42:35Z","timestamp":1780321355797,"version":"3.54.1"},"reference-count":76,"publisher":"Cambridge University Press (CUP)","issue":"2","license":[{"start":{"date-parts":[[2022,10,18]],"date-time":"2022-10-18T00:00:00Z","timestamp":1666051200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/www.cambridge.org\/core\/terms"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Econom. Theory"],"published-print":{"date-parts":[[2024,4]]},"abstract":"<jats:p>Regularized quantile regression (QR) is a useful technique for analyzing heterogeneous data under potentially heavy-tailed error contamination in high dimensions. This paper provides a new analysis of the estimation\/prediction error bounds of the global solution of<jats:inline-formula><jats:alternatives><jats:inline-graphic xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" mime-subtype=\"png\" xlink:href=\"S0266466622000421_inline1.png\"\/><jats:tex-math>$L_1$<\/jats:tex-math><\/jats:alternatives><\/jats:inline-formula>-regularized QR (QR-LASSO) and the local solutions of nonconvex regularized QR (QR-NCP) when the number of covariates is greater than the sample size. Our results build upon and significantly generalize the earlier work in the literature. For certain heavy-tailed error distributions and a general class of design matrices, the least-squares-based LASSO cannot achieve the near-oracle rate derived under the normality assumption no matter the choice of the tuning parameter. In contrast, we establish that QR-LASSO achieves the near-oracle estimation error rate for a broad class of models under conditions weaker than those in the literature. For QR-NCP, we establish the novel results that all local optima within a feasible region have desirable estimation accuracy. Our analysis applies to not just the hard sparsity setting commonly used in the literature, but also to the soft sparsity setting which permits many small coefficients. Our approach relies on a unified characterization of the global\/local solutions of regularized QR via subgradients using a generalized Karush\u2013Kuhn\u2013Tucker condition. The theory of the paper establishes a key property of the subdifferential of the quantile loss function in high dimensions, which is of independent interest for analyzing other high-dimensional nonsmooth problems.<\/jats:p>","DOI":"10.1017\/s0266466622000421","type":"journal-article","created":{"date-parts":[[2022,10,18]],"date-time":"2022-10-18T07:18:07Z","timestamp":1666077487000},"page":"233-277","source":"Crossref","is-referenced-by-count":13,"title":["ANALYSIS OF GLOBAL AND LOCAL OPTIMA OF REGULARIZED QUANTILE REGRESSION IN HIGH DIMENSIONS: A SUBGRADIENT APPROACH"],"prefix":"10.1017","volume":"40","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3217-0202","authenticated-orcid":false,"given":"Lan","family":"Wang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3442-4173","authenticated-orcid":false,"given":"Xuming","family":"He","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"56","published-online":{"date-parts":[[2022,10,18]]},"reference":[{"key":"S0266466622000421_r15","doi-asserted-by":"publisher","DOI":"10.1214\/18-AOS1777"},{"key":"S0266466622000421_r67","unstructured":"Wang, L. (2019). L1 -regularized quantile regression with many regressors under lean assumptions. University of Minnesota Digital Conservancy. Available at https:\/\/hdl.handle.net\/11299\/202063."},{"key":"S0266466622000421_r43","unstructured":"Linton, O.B. & Whang, Y.-J. (2004). A quantilogram approach to evaluating directional predictability. Available at SSRN 485342."},{"key":"S0266466622000421_r14","doi-asserted-by":"crossref","first-page":"433","DOI":"10.1016\/j.jeconom.2019.04.037","article-title":"Nonparametric estimation of conditional quantile functions in the presence of irrelevant covariates","volume":"212","author":"Chen","year":"2019a","journal-title":"Journal of Econometrics"},{"key":"S0266466622000421_r42","doi-asserted-by":"publisher","DOI":"10.1198\/106186008X289155"},{"key":"S0266466622000421_r12","doi-asserted-by":"publisher","DOI":"10.1214\/07-EJS008"},{"key":"S0266466622000421_r73","unstructured":"Zhao, T. , Kolar, M. , & Liu, H. (2014) A general framework for robust testing and confidence regions in high-dimensional quantile regression. Preprint, arXiv:1412.8724."},{"key":"S0266466622000421_r27","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2018.06.004"},{"key":"S0266466622000421_r6","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2018.1442339"},{"key":"S0266466622000421_r3","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA14030"},{"key":"S0266466622000421_r13","doi-asserted-by":"publisher","DOI":"10.1017\/CCOL0521444594.005"},{"key":"S0266466622000421_r30","doi-asserted-by":"publisher","DOI":"10.3150\/18-BEJ1091"},{"key":"S0266466622000421_r66","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmva.2013.04.001"},{"key":"S0266466622000421_r70","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2012.656014"},{"key":"S0266466622000421_r28","doi-asserted-by":"publisher","DOI":"10.3150\/bj\/1106314846"},{"key":"S0266466622000421_r50","doi-asserted-by":"publisher","DOI":"10.21236\/ADA415451"},{"key":"S0266466622000421_r23","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1016\/j.jmva.2017.11.001","article-title":"Quantile regression for additive coefficient models in high dimensions","volume":"164","author":"Fan","year":"2018","journal-title":"Journal of Multivariate Analysis"},{"key":"S0266466622000421_r62","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4757-2545-2"},{"key":"S0266466622000421_r9","unstructured":"Bradic, J. & Kolar, M. (2017). Uniform inference for high-dimensional quantile regression: Linear functionals and regression rank scores. Preprint, arXiv:1702.06209."},{"key":"S0266466622000421_r10","doi-asserted-by":"publisher","DOI":"10.2307\/2951618"},{"key":"S0266466622000421_r34","unstructured":"Kato, K. (2011) Group Lasso for high dimensional sparse quantile regression models. Preprint, arXiv:1103.1458."},{"key":"S0266466622000421_r32","doi-asserted-by":"publisher","DOI":"10.1198\/016214502388618627"},{"key":"S0266466622000421_r19","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2018.2863700"},{"key":"S0266466622000421_r26","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2013.818002"},{"key":"S0266466622000421_r35","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-economics-063016-103651"},{"key":"S0266466622000421_r11","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1099-1255(199801\/02)13:1<1::AID-JAE474>3.0.CO;2-A"},{"key":"S0266466622000421_r48","doi-asserted-by":"publisher","DOI":"10.1214\/17-AOS1637"},{"key":"S0266466622000421_r58","first-page":"289","article-title":"Convex analysis approach to D.C. programming: Theory, algorithms and applications","volume":"22","author":"Tao","year":"1997","journal-title":"Acta Mathematica Vietnamica"},{"key":"S0266466622000421_r52","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2011.2165799"},{"key":"S0266466622000421_r63","doi-asserted-by":"publisher","DOI":"10.3103\/S1066530712020044"},{"key":"S0266466622000421_r31","doi-asserted-by":"publisher","DOI":"10.1198\/016214505000000583"},{"key":"S0266466622000421_r72","doi-asserted-by":"publisher","DOI":"10.1214\/09-AOS729"},{"key":"S0266466622000421_r55","doi-asserted-by":"publisher","DOI":"10.1016\/j.jspi.2010.01.043"},{"key":"S0266466622000421_r39","volume-title":"Probability in Banach Spaces: Isoperimetry and Processes","author":"Ledoux","year":"2013"},{"key":"S0266466622000421_r5","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/asu056"},{"key":"S0266466622000421_r8","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-9868.2010.00764.x"},{"key":"S0266466622000421_r76","doi-asserted-by":"publisher","DOI":"10.1214\/07-AOS507"},{"key":"S0266466622000421_r24","doi-asserted-by":"crossref","first-page":"953","DOI":"10.3982\/ECTA6822","article-title":"Unconditional quantile regressions","volume":"77","author":"Firpo","year":"2009","journal-title":"Econometrica"},{"key":"S0266466622000421_r36","doi-asserted-by":"publisher","DOI":"10.2307\/1913643"},{"key":"S0266466622000421_r25","volume-title":"Economic Applications of Quantile Regression","author":"Fitzenberger","year":"2013"},{"key":"S0266466622000421_r65","first-page":"145","article-title":"Variable selection for censored quantile regression","volume":"23","author":"Wang","year":"2013a","journal-title":"Statistica Sinica"},{"key":"S0266466622000421_r59","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.2517-6161.1996.tb02080.x","article-title":"Regression shrinkage and selection via the Lasso","volume":"58","author":"Tibshirani","year":"1996","journal-title":"Journal of the Royal Statistical Society. Series B"},{"key":"S0266466622000421_r21","doi-asserted-by":"publisher","DOI":"10.1111\/rssb.12166"},{"key":"S0266466622000421_r53","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1980.10477560"},{"key":"S0266466622000421_r33","doi-asserted-by":"publisher","DOI":"10.1214\/10-AOS842"},{"key":"S0266466622000421_r64","doi-asserted-by":"publisher","DOI":"10.1198\/073500106000000251"},{"key":"S0266466622000421_r69","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2020.1840989"},{"key":"S0266466622000421_r7","doi-asserted-by":"publisher","DOI":"10.1214\/08-AOS620"},{"key":"S0266466622000421_r44","doi-asserted-by":"publisher","DOI":"10.1214\/16-AOS1471"},{"key":"S0266466622000421_r49","doi-asserted-by":"publisher","DOI":"10.1214\/12-STS400"},{"key":"S0266466622000421_r17","doi-asserted-by":"crossref","first-page":"535","DOI":"10.3982\/ECTA8405","article-title":"Average and quantile effects in nonseparable panel models","volume":"81","author":"Chernozhukov","year":"2013","journal-title":"Econometrica"},{"key":"S0266466622000421_r60","volume-title":"Empirical Processes in M-Estimation","author":"van de Geer","year":"2000"},{"key":"S0266466622000421_r41","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2017.1319840"},{"key":"S0266466622000421_r37","doi-asserted-by":"publisher","DOI":"10.1201\/9781315120256"},{"key":"S0266466622000421_r57","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmva.2013.07.015"},{"key":"S0266466622000421_r68","doi-asserted-by":"publisher","DOI":"10.1214\/13-AOS1159"},{"key":"S0266466622000421_r71","first-page":"801","article-title":"Variable selection in quantile regression","volume":"19","author":"Wu","year":"2009","journal-title":"Statistica Sinica"},{"key":"S0266466622000421_r74","doi-asserted-by":"publisher","DOI":"10.1214\/15-AOS1340"},{"key":"S0266466622000421_r51","first-page":"1619","article-title":"Dantzig-type penalization for multiple quantile regression with high dimensional covariates","volume":"27","author":"Park","year":"2017","journal-title":"Statistica Sinica"},{"key":"S0266466622000421_r29","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2018.12.006"},{"key":"S0266466622000421_r40","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2013.836975"},{"key":"S0266466622000421_r45","doi-asserted-by":"publisher","DOI":"10.1214\/12-AOS1018"},{"key":"S0266466622000421_r1","doi-asserted-by":"publisher","DOI":"10.1111\/1468-0262.00270"},{"key":"S0266466622000421_r18","doi-asserted-by":"publisher","DOI":"10.1007\/BF01199026"},{"key":"S0266466622000421_r22","doi-asserted-by":"publisher","DOI":"10.1198\/016214501753382273"},{"key":"S0266466622000421_r54","doi-asserted-by":"crossref","first-page":"288","DOI":"10.1214\/15-AOS1367","article-title":"Partially linear additive quantile regression in ultra-high dimension","volume":"44","author":"Sherwood","year":"2016","journal-title":"Annals of Statistics"},{"key":"S0266466622000421_r46","first-page":"559","article-title":"Regularized $m$-estimators with nonconvexity: Statistical and algorithmic theory for local optima","volume":"16","author":"Loh","year":"2015","journal-title":"Journal of Machine Learning Research"},{"key":"S0266466622000421_r20","doi-asserted-by":"publisher","DOI":"10.1214\/13-AOS1191"},{"key":"S0266466622000421_r4","doi-asserted-by":"publisher","DOI":"10.1214\/10-AOS827"},{"key":"S0266466622000421_r16","doi-asserted-by":"publisher","DOI":"10.1093\/restud\/rdq020"},{"key":"S0266466622000421_r38","doi-asserted-by":"publisher","DOI":"10.1198\/016214506000000672"},{"key":"S0266466622000421_r47","doi-asserted-by":"publisher","DOI":"10.1214\/17-AOS1567"},{"key":"S0266466622000421_r56","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2015.10.006"},{"key":"S0266466622000421_r75","first-page":"69","article-title":"Regularized quantile regression and robust feature screening for single index models","volume":"26","author":"Zhong","year":"2016","journal-title":"Statistica Sinica"},{"key":"S0266466622000421_r61","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-32774-7"},{"key":"S0266466622000421_r2","doi-asserted-by":"publisher","DOI":"10.1111\/j.1468-0262.2006.00671.x"}],"container-title":["Econometric Theory"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.cambridge.org\/core\/services\/aop-cambridge-core\/content\/view\/S0266466622000421","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,10,6]],"date-time":"2024-10-06T00:05:29Z","timestamp":1728173129000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.cambridge.org\/core\/product\/identifier\/S0266466622000421\/type\/journal_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,18]]},"references-count":76,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2024,4]]}},"alternative-id":["S0266466622000421"],"URL":"https:\/\/doi.org\/10.1017\/s0266466622000421","relation":{},"ISSN":["0266-4666","1469-4360"],"issn-type":[{"value":"0266-4666","type":"print"},{"value":"1469-4360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,10,18]]}}}