{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,15]],"date-time":"2026-02-15T03:15:36Z","timestamp":1771125336071,"version":"3.50.1"},"reference-count":51,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2022,2,16]],"date-time":"2022-02-16T00:00:00Z","timestamp":1644969600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,11,29]]},"abstract":"<jats:title>Summary<\/jats:title><jats:p>This paper addresses the task of estimating a covariance matrix under a patternless sparsity assumption. In contrast to existing approaches based on thresholding or shrinkage penalties, we propose a likelihood-based method that regularizes the distance from the covariance estimate to a symmetric sparsity set. This formulation avoids unwanted shrinkage induced by more common norm penalties, and enables optimization of the resulting nonconvex objective by solving a sequence of smooth, unconstrained subproblems. These subproblems are generated and solved via the proximal distance version of the majorization-minimization principle. The resulting algorithm executes rapidly, gracefully handles settings where the number of parameters exceeds the number of cases, yields a positive-definite solution, and enjoys desirable convergence properties. Empirically, we demonstrate that our approach outperforms competing methods across several metrics, for a suite of simulated experiments. Its merits are illustrated on international migration data and a case study on flow cytometry. Our findings suggest that the marginal and conditional dependency networks for the cell signalling data are more similar than previously concluded.<\/jats:p>","DOI":"10.1093\/biomet\/asac011","type":"journal-article","created":{"date-parts":[[2022,2,9]],"date-time":"2022-02-09T12:22:52Z","timestamp":1644409372000},"page":"1047-1066","source":"Crossref","is-referenced-by-count":15,"title":["A proximal distance algorithm for likelihood-based sparse covariance estimation"],"prefix":"10.1093","volume":"109","author":[{"given":"Jason","family":"Xu","sequence":"first","affiliation":[{"name":"Duke University Department of Statistical Science, , Box 90251, Durham, North Carolina 27708, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kenneth","family":"Lange","sequence":"additional","affiliation":[{"name":"University of California Department of Computational Medicine, , Los Angeles, Box 708822, Los Angeles, California 90095, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2022,2,16]]},"reference":[{"key":"2022112917593490500_B1","doi-asserted-by":"crossref","first-page":"940","DOI":"10.1214\/18-AOAS1175","article-title":"Estimating large correlation matrices for international migration","volume":"12","author":"Azose,","year":"2018","journal-title":"Ann. Appl. Statist."},{"key":"2022112917593490500_B2","doi-asserted-by":"crossref","first-page":"820","DOI":"10.1145\/361573.361582","article-title":"Solution of the matrix equation AX + XB = C","volume":"15","author":"Bartels,","year":"1972","journal-title":"Commun. ACM"},{"key":"2022112917593490500_B3","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4419-9467-7","volume-title":"Convex Analysis and Monotone Operator Theory in Hilbert Spaces","author":"Bauschke,","year":"2011"},{"key":"2022112917593490500_B4","doi-asserted-by":"crossref","DOI":"10.1137\/1.9781611974997","volume-title":"First-order Methods in Optimization","author":"Beck,","year":"2017"},{"key":"2022112917593490500_B5","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1137\/080716542","article-title":"A fast iterative shrinkage-thresholding algorithm for linear inverse problems","volume":"2","author":"Beck,","year":"2009","journal-title":"SIAM J. Imag. Sci."},{"key":"2022112917593490500_B6","doi-asserted-by":"crossref","first-page":"2577","DOI":"10.1214\/08-AOS600","article-title":"Covariance regularization by thresholding","volume":"36","author":"Bickel,","year":"2008","journal-title":"Ann. Statist."},{"key":"2022112917593490500_B7","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1214\/009053607000000758","article-title":"Regularized estimation of large covariance matrices","volume":"36","author":"Bickel,","year":"2008","journal-title":"Ann. Statist."},{"key":"2022112917593490500_B8","doi-asserted-by":"crossref","first-page":"834","DOI":"10.1080\/01621459.2015.1058265","article-title":"Convex banding of the covariance matrix","volume":"111","author":"Bien,","year":"2016","journal-title":"J. Am. Statist. Assoc."},{"key":"2022112917593490500_B9","doi-asserted-by":"crossref","first-page":"807","DOI":"10.1093\/biomet\/asr054","article-title":"Sparse estimation of a covariance matrix","volume":"98","author":"Bien,","year":"2011","journal-title":"Biometrika"},{"key":"2022112917593490500_B10","doi-asserted-by":"crossref","first-page":"775","DOI":"10.1111\/j.1467-985X.2009.00635.x","article-title":"Bayesian forecasting of immigration to selected European countries by using expert knowledge","volume":"173","author":"Bijak,","year":"2010","journal-title":"J. R. Statist. Soc. A"},{"key":"2022112917593490500_B11","first-page":"1","article-title":"Distributed optimization and statistical learning via the alternating direction method of multipliers","volume":"3","author":"Boyd,","year":"2011","journal-title":"Foundat. Trends Mach. Learn."},{"key":"2022112917593490500_B12","doi-asserted-by":"crossref","first-page":"672","DOI":"10.1198\/jasa.2011.tm10560","article-title":"Adaptive thresholding for sparse covariance matrix estimation","volume":"106","author":"Cai,","year":"2011","journal-title":"J. Am. Statist. Assoc."},{"key":"2022112917593490500_B13","doi-asserted-by":"crossref","first-page":"2118","DOI":"10.1214\/09-AOS752","article-title":"Optimal rates of convergence for covariance matrix estimation","volume":"38","author":"Cai,","year":"2010","journal-title":"Ann. Statist."},{"key":"2022112917593490500_B14","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1093\/biomet\/asm007","article-title":"Estimation of a covariance matrix with zeros","volume":"94","author":"Chaudhuri,","year":"2007","journal-title":"Biometrika"},{"key":"2022112917593490500_B15","doi-asserted-by":"crossref","first-page":"117","DOI":"10.1016\/j.csda.2014.06.018","article-title":"Stable estimation of a covariance matrix guided by nuclear norm penalties","volume":"80","author":"Chi,","year":"2014","journal-title":"Comp. Statist. Data Anal."},{"key":"2022112917593490500_B16","doi-asserted-by":"crossref","first-page":"409","DOI":"10.1007\/s10107-013-0697-1","article-title":"Distance majorization and its applications","volume":"146","author":"Chi,","year":"2014","journal-title":"Math. Program."},{"key":"2022112917593490500_B17","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1090\/S0002-9904-1943-07818-4","article-title":"Variational methods for the solution of problems of equilibrium and vibrations","volume":"49","author":"Courant,","year":"1943","journal-title":"Bull. Am. Math. Soc."},{"key":"2022112917593490500_B18","doi-asserted-by":"crossref","first-page":"390","DOI":"10.1016\/j.csda.2014.10.001","article-title":"Sparse estimation of high-dimensional correlation matrices","volume":"93","author":"Cui,","year":"2016","journal-title":"Comp. Statist. Data Anal."},{"key":"2022112917593490500_B19","doi-asserted-by":"crossref","first-page":"3344","DOI":"10.1137\/18M117337X","article-title":"Composite difference-max programs for modern statistical estimation problems","volume":"28","author":"Cui,","year":"2018","journal-title":"SIAM J. Optimiz."},{"key":"2022112917593490500_B20","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","article-title":"Maximum likelihood from incomplete data via the EM algorithm","volume":"39","author":"Dempster,","year":"1977","journal-title":"J. R. Statist. Soc. B"},{"key":"2022112917593490500_B21","doi-asserted-by":"crossref","first-page":"C1","DOI":"10.1111\/ectj.12061","article-title":"An overview of the estimation of large covariance and precision matrices","volume":"19","author":"Fan,","year":"2016","journal-title":"Economet. J."},{"key":"2022112917593490500_B22","doi-asserted-by":"crossref","first-page":"494","DOI":"10.1080\/00949655.2015.1017823","article-title":"Tuning-parameter selection in regularized estimations of large covariance matrices","volume":"86","author":"Fang,","year":"2016","journal-title":"J. Statist. Comp. Simul."},{"key":"2022112917593490500_B23","doi-asserted-by":"crossref","first-page":"432","DOI":"10.1093\/biostatistics\/kxm045","article-title":"Sparse inverse covariance estimation with the graphical lasso","volume":"9","author":"Friedman,","year":"2008","journal-title":"Biostatistics"},{"key":"2022112917593490500_B24","doi-asserted-by":"crossref","DOI":"10.1137\/1.9780898718027","volume-title":"Accuracy and Stability of Numerical Algorithms","author":"Higham,","year":"2002"},{"key":"2022112917593490500_B25","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1093\/biomet\/93.1.85","article-title":"Covariance matrix selection and estimation via penalised normal likelihood","volume":"93","author":"Huang,","year":"2006","journal-title":"Biometrika"},{"key":"2022112917593490500_B26","first-page":"2717","article-title":"Operator norm consistent estimation of large-dimensional sparse covariance matrices","volume":"36","author":"Karoui,","year":"2008","journal-title":"Ann. Statist."},{"key":"2022112917593490500_B27","first-page":"1","article-title":"Proximal distance algorithms: Theory and practice","volume":"20","author":"Keys,","year":"2019","journal-title":"J. Mach. Learn. Res."},{"key":"2022112917593490500_B28","doi-asserted-by":"crossref","first-page":"4254","DOI":"10.1214\/09-AOS720","article-title":"Sparsistency and rates of convergence in large covariance matrix estimation","volume":"37","author":"Lam,","year":"2009","journal-title":"Ann. Statist."},{"key":"2022112917593490500_B29","doi-asserted-by":"crossref","DOI":"10.1137\/1.9781611974409","volume-title":"MM Optimization Algorithms","author":"Lange,","year":"2016"},{"key":"2022112917593490500_B30","doi-asserted-by":"crossref","first-page":"245","DOI":"10.1214\/07-AOAS139","article-title":"Sparse estimation of large covariance matrices via a nested lasso penalty","volume":"2","author":"Levina,","year":"2008","journal-title":"Ann. Appl. Statist."},{"key":"2022112917593490500_B31","volume-title":"Linear and Nonlinear Programming","author":"Luenberger,","year":"1984"},{"key":"2022112917593490500_B32","doi-asserted-by":"crossref","first-page":"829","DOI":"10.1137\/140957639","article-title":"Incremental majorization-minimization optimization with application to large-scale machine learning","volume":"25","author":"Mairal,","year":"2015","journal-title":"SIAM J. Optimiz."},{"key":"2022112917593490500_B33","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1093\/biomet\/asy023","article-title":"Shrinking characteristics of precision matrix estimators","volume":"105","author":"Molstad,","year":"2018","journal-title":"Biometrika"},{"key":"2022112917593490500_B34","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1287\/moor.2016.0795","article-title":"Computing B-stationary points of nonsmooth DC programs","volume":"42","author":"Pang,","year":"2017","journal-title":"Math. Oper. Res."},{"key":"2022112917593490500_B35","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1214\/15-STS530","article-title":"Proximal algorithms in statistics and machine learning","volume":"30","author":"Polson,","year":"2015","journal-title":"Statist. Sci."},{"key":"2022112917593490500_B36","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1214\/11-STS358","article-title":"Covariance estimation: The GLM and regularization perspectives","volume":"26","author":"Pourahmadi,","year":"2011","journal-title":"Statist. Sci."},{"key":"2022112917593490500_B37","volume-title":"R: A Language and Environment for Statistical Computing","year":"2022"},{"key":"2022112917593490500_B38","doi-asserted-by":"crossref","first-page":"733","DOI":"10.1093\/biomet\/ass025","article-title":"Positive definite estimators of large covariance matrices","volume":"99","author":"Rothman,","year":"2012","journal-title":"Biometrika"},{"key":"2022112917593490500_B39","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1198\/jasa.2009.0101","article-title":"Generalized thresholding of large covariance matrices","volume":"104","author":"Rothman,","year":"2009","journal-title":"J. Am. Statist. Assoc."},{"key":"2022112917593490500_B40","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1126\/science.1105809","article-title":"Causal protein-signaling networks derived from multiparameter single-cell data","volume":"308","author":"Sachs,","year":"2005","journal-title":"Science"},{"key":"2022112917593490500_B41","doi-asserted-by":"crossref","first-page":"377","DOI":"10.1137\/130912839","article-title":"Computational methods for linear matrix equations","volume":"58","author":"Simoncini,","year":"2016","journal-title":"SIAM Review"},{"key":"2022112917593490500_B42","first-page":"197","article-title":"Inadmissibility of the usual estimator for the mean of a multivariate normal distribution","volume-title":"Proc. 3rd Berkeley Symp. Math. Statist. Prob.","author":"Stein,","year":"1956"},{"key":"2022112917593490500_B43","first-page":"3642","article-title":"Projection onto Minkowski sums with application to constrained learning","volume":"97","author":"Won,","year":"2019","journal-title":"Proc. Mach. Learn. Res."},{"key":"2022112917593490500_B44","doi-asserted-by":"crossref","first-page":"831","DOI":"10.1093\/biomet\/90.4.831","article-title":"Nonparametric estimation of large covariance matrices of longitudinal data","volume":"90","author":"Wu,","year":"2003","journal-title":"Biometrika"},{"key":"2022112917593490500_B45","first-page":"1385","article-title":"Generalized linear model regression under distance-to-set penalties","volume-title":"Proc. 31st Int. Conf. Neural Information Processing Systems (NIPS\u201917)","author":"Xu,","year":"2017"},{"key":"2022112917593490500_B46","first-page":"6921","article-title":"Power k-means clustering","volume":"97","author":"Xu,","year":"2019","journal-title":"Proc. Mach. Learn. Res."},{"key":"2022112917593490500_B47","doi-asserted-by":"crossref","first-page":"1480","DOI":"10.1080\/01621459.2012.725386","article-title":"Positive-definite $$\\ell_1$$-penalized estimation of large covariance matrices","volume":"107","author":"Xue,","year":"2012","journal-title":"J. Am. Statist. Assoc."},{"key":"2022112917593490500_B48","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1093\/biomet\/asm018","article-title":"Model selection and estimation in the Gaussian graphical model","volume":"94","author":"Yuan,","year":"2007","journal-title":"Biometrika"},{"key":"2022112917593490500_B49","doi-asserted-by":"crossref","first-page":"915","DOI":"10.1162\/08997660360581958","article-title":"The concave-convex procedure","volume":"15","author":"Yuille,","year":"2003","journal-title":"Neural Comp."},{"key":"2022112917593490500_B50","volume-title":"Nonlinear Programming: A Unified Approach","author":"Zangwill,","year":"1969"},{"key":"2022112917593490500_B51","first-page":"1509","article-title":"One-step sparse estimates in nonconcave penalized likelihood models","volume":"36","author":"Zou,","year":"2008","journal-title":"Ann. Statist."}],"container-title":["Biometrika"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/biomet\/advance-article-pdf\/doi\/10.1093\/biomet\/asac011\/43540283\/asac011.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/109\/4\/1047\/47415669\/asac011.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/109\/4\/1047\/47415669\/asac011.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,18]],"date-time":"2024-09-18T04:53:26Z","timestamp":1726635206000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biomet\/article\/109\/4\/1047\/6529114"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2,16]]},"references-count":51,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2022,2,16]]},"published-print":{"date-parts":[[2022,11,29]]}},"URL":"https:\/\/doi.org\/10.1093\/biomet\/asac011","relation":{},"ISSN":["0006-3444","1464-3510"],"issn-type":[{"value":"0006-3444","type":"print"},{"value":"1464-3510","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,12,1]]},"published":{"date-parts":[[2022,2,16]]}}}