{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T19:58:46Z","timestamp":1791403126718,"version":"4.3.3"},"reference-count":35,"publisher":"Institute of Mathematical Statistics","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Ann. Statist."],"published-print":{"date-parts":[[2022,6,1]]},"abstract":"<jats:p>Given n independent random vectors with common density f on Rd, we study the weak convergence of three empirical-measure based estimators of the convex \u03bb-level set L\u03bb of f, namely the excess mass set, the minimum volume set and the maximum probability set, all selected from a class of convex sets A that contains L\u03bb. Since these set-valued estimators approach L\u03bb, even the formulation of their weak convergence is nonstandard. We identify the joint limiting distribution of the symmetric difference of L\u03bb and each of the three estimators, at rate n\u22121\/3. It turns out that the minimum volume set and the maximum probability set estimators are asymptotically indistinguishable, whereas the excess mass set estimator exhibits \u201cricher\u201d limit behavior. Arguments rely on the boundary local empirical process, its cylinder representation, dimension-free concentration around the boundary of L\u03bb, and the set-valued argmax of a drifted Wiener process.<\/jats:p>","DOI":"10.1214\/21-aos2157","type":"journal-article","created":{"date-parts":[[2022,6,17]],"date-time":"2022-06-17T01:51:24Z","timestamp":1655430684000},"page":"1423-1446","source":"Crossref","is-referenced-by-count":1,"title":["Cube root weak convergence of empirical estimators of a density level set"],"prefix":"10.1214","volume":"50","author":[{"given":"Philippe","family":"Berthet","sequence":"first","affiliation":[{"name":"Institut de Math\u00e9matiques de Toulouse; UMR5219, Universit\u00e9 de Toulouse; CNRS"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John H. J.","family":"Einmahl","sequence":"additional","affiliation":[{"name":"Department of Econometrics and OR and CentER, Tilburg University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"108","reference":[{"key":"29","doi-asserted-by":"publisher","unstructured":"<small>Schneider, R.<\/small> and <small>Weil, W.<\/small> (2008). <i>Stochastic and Integral Geometry<\/i>. <i>Probability and Its Applications<\/i> (<i>New York<\/i>). Springer, Berlin.","DOI":"10.1007\/978-3-540-78859-1"},{"key":"22","doi-asserted-by":"publisher","unstructured":"<small>Polonik, W.<\/small> (1995). Measuring mass concentrations and estimating density contour clusters\u2014an excess mass approach. <i>Ann. Statist.<\/i> <b>23<\/b> 855\u2013881.","DOI":"10.1214\/aos\/1176324626"},{"key":"23","doi-asserted-by":"publisher","unstructured":"<small>Polonik, W.<\/small> (1997). Minimum volume sets and generalized quantile processes. <i>Stochastic Process. Appl.<\/i> <b>69<\/b> 1\u201324.","DOI":"10.1016\/S0304-4149(97)00028-8"},{"key":"27","doi-asserted-by":"crossref","unstructured":"<small>ROUSSEEUW, P.<\/small> (1985). Multivariate estimation with high breakdown point. In <i>Mathematical Statistics and Applications, Vol. B<\/i> (<i>Bad Tatzmannsdorf<\/i>, 1983) 283\u2013297. Reidel, Dordrecht.","DOI":"10.1007\/978-94-009-5438-0_20"},{"key":"1","doi-asserted-by":"crossref","unstructured":"<small>ALEXANDER, K. S.<\/small> (1984). Probability inequalities for empirical processes and a law of the iterated logarithm. <i>Ann. Probab.<\/i> <b>12<\/b> 1041\u20131067.","DOI":"10.1214\/aop\/1176993141"},{"key":"2","unstructured":"<small>BERTHET, P.<\/small> and <small>EL-NOUTY, C.<\/small> (2006). Almost sure asymptotic behaviour of the shorth estimators. <i>Math. Methods Statist.<\/i> <b>14<\/b> 379\u2013403."},{"key":"4","doi-asserted-by":"publisher","unstructured":"<small>CADRE, B.<\/small> (2006). Kernel estimation of density level sets. <i>J. Multivariate Anal.<\/i> <b>97<\/b> 999\u20131023.","DOI":"10.1016\/j.jmva.2005.05.004"},{"key":"5","doi-asserted-by":"publisher","unstructured":"<small>CAI, J.-J.<\/small>, <small>EINMAHL, J. H. J.<\/small> and <small>DE HAAN, L.<\/small> (2011). Estimation of extreme risk regions under multivariate regular variation. <i>Ann. Statist.<\/i> <b>39<\/b> 1803\u20131826.","DOI":"10.1214\/11-AOS891"},{"key":"6","doi-asserted-by":"publisher","unstructured":"<small>CHEN, Y.-C.<\/small>, <small>GENOVESE, C. R.<\/small> and <small>WASSERMAN, L.<\/small> (2017). Density level sets: Asymptotics, inference, and visualization. <i>J. Amer. Statist. Assoc.<\/i> <b>112<\/b> 1684\u20131696.","DOI":"10.1080\/01621459.2016.1228536"},{"key":"7","unstructured":"<small>CL\u00c9MEN\u00c7ON, S.<\/small>, <small>GOIX, N.<\/small> and <small>SABOURIN, A.<\/small> (2015). On anomaly ranking and excess-mass curves. In <i>Proceedings of the AISTATS<\/i> 2015 287\u2013295. San Diego."},{"key":"8","doi-asserted-by":"publisher","unstructured":"<small>DAVIES, L.<\/small> (1992). The asymptotics of Rousseeuw\u2019s minimum volume ellipsoid estimator. <i>Ann. Statist.<\/i> <b>20<\/b> 1828\u20131843.","DOI":"10.1214\/aos\/1176348891"},{"key":"9","doi-asserted-by":"publisher","unstructured":"<small>DI BUCCHIANICO, A.<\/small>, <small>EINMAHL, J. H. J.<\/small> and <small>MUSHKUDIANI, N. A.<\/small> (2001). Smallest nonparametric tolerance regions. <i>Ann. Statist.<\/i> <b>29<\/b> 1320\u20131343.","DOI":"10.1214\/aos\/1013203456"},{"key":"10","doi-asserted-by":"publisher","unstructured":"<small>DYKSTRA, R.<\/small> and <small>CAROLAN, C.<\/small> (1999). The distribution of the argmax of two-sided Brownian motion with quadratic drift. <i>J. Statist. Comput. Simulation<\/i> <b>63<\/b> 47\u201358.","DOI":"10.1080\/00949659908811948"},{"key":"11","doi-asserted-by":"publisher","unstructured":"<small>EINMAHL, J. H. J.<\/small> and <small>KHMALADZE, E. V.<\/small> (2011). Central limit theorems for local empirical processes near boundaries of sets. <i>Bernoulli<\/i> <b>17<\/b> 545\u2013561.","DOI":"10.3150\/10-BEJ283"},{"key":"13","doi-asserted-by":"crossref","unstructured":"<small>HARTIGAN, J. A.<\/small> (1987). Estimation of a convex density contour in two dimensions. <i>J. Amer. Statist. Assoc.<\/i> <b>82<\/b> 267\u2013270.","DOI":"10.1080\/01621459.1987.10478428"},{"key":"14","doi-asserted-by":"publisher","unstructured":"<small>KHMALADZE, E. V.<\/small> (2007). Differentiation of sets in measure. <i>J. Math. Anal. Appl.<\/i> <b>334<\/b> 1055\u20131072.","DOI":"10.1016\/j.jmaa.2006.12.086"},{"key":"15","doi-asserted-by":"publisher","unstructured":"<small>KHMALADZE, E.<\/small> and <small>WEIL, W.<\/small> (2008). Local empirical processes near boundaries of convex bodies. <i>Ann. Inst. Statist. Math.<\/i> <b>60<\/b> 813\u2013842.","DOI":"10.1007\/s10463-007-0123-7"},{"key":"17","doi-asserted-by":"publisher","unstructured":"<small>MAMMEN, E.<\/small> and <small>POLONIK, W.<\/small> (2013). Confidence regions for level sets. <i>J. Multivariate Anal.<\/i> <b>122<\/b> 202\u2013214.","DOI":"10.1016\/j.jmva.2013.07.017"},{"key":"18","doi-asserted-by":"publisher","unstructured":"<small>MASON, D. M.<\/small> and <small>POLONIK, W.<\/small> (2009). Asymptotic normality of plug-in level set estimates. <i>Ann. Appl. Probab.<\/i> <b>19<\/b> 1108\u20131142.","DOI":"10.1214\/08-AAP569"},{"key":"19","unstructured":"<small>M\u00dcLLER, D. W.<\/small> (1992). <i>The Excess Mass Approach in Statistics<\/i>. Beitr\u00e4ge zur Statistik, Univ. Heidelberg."},{"key":"20","doi-asserted-by":"crossref","unstructured":"<small>M\u00dcLLER, D. W.<\/small> and <small>SAWITZKI, G.<\/small> (1991). Excess mass estimates and tests for multimodality. <i>J. Amer. Statist. Assoc.<\/i> <b>86<\/b> 738\u2013746.","DOI":"10.1080\/01621459.1991.10475103"},{"key":"21","doi-asserted-by":"publisher","unstructured":"<small>NOLAN, D.<\/small> (1991). The excess-mass ellipsoid. <i>J. Multivariate Anal.<\/i> <b>39<\/b> 348\u2013371.","DOI":"10.1016\/0047-259X(91)90106-C"},{"key":"25","doi-asserted-by":"publisher","unstructured":"<small>QIAO, W.<\/small> and <small>POLONIK, W.<\/small> (2019). Nonparametric confidence regions for level sets: Statistical properties and geometry. <i>Electron. J. Stat.<\/i> <b>13<\/b> 985\u20131030.","DOI":"10.1214\/19-EJS1543"},{"key":"26","unstructured":"<small>RODR\u00cdGUEZ-CASAL, A.<\/small> and <small>SAAVEDRA-NIEVES, P.<\/small> (2019). Minimax Hausdorff estimation of density level sets. Preprint. Available at <a href=\"https:\/\/arxiv.org\/abs\/arXiv:1905.02897\">arXiv:1905.02897<\/a>."},{"key":"28","doi-asserted-by":"publisher","unstructured":"<small>SCHNEIDER, R.<\/small> (1993). <i>Convex Bodies<\/i>: <i>The Brunn-Minkowski Theory<\/i>. <i>Encyclopedia of Mathematics and Its Applications<\/i> <b>44<\/b>. Cambridge Univ. Press, Cambridge.","DOI":"10.1017\/CBO9780511526282"},{"key":"30","unstructured":"<small>SCOTT, C. D.<\/small> and <small>NOWAK, R. D.<\/small> (2006). Learning minimum volume sets. <i>J. Mach. Learn. Res.<\/i> <b>7<\/b> 665\u2013704."},{"key":"31","doi-asserted-by":"publisher","unstructured":"<small>SILVERMAN, B. W.<\/small> and <small>TITTERINGTON, D. M.<\/small> (1980). Minimum covering ellipses. <i>SIAM J. Sci. Statist. Comput.<\/i> <b>1<\/b> 401\u2013409.","DOI":"10.1137\/0901028"},{"key":"32","doi-asserted-by":"publisher","unstructured":"<small>TSYBAKOV, A. B.<\/small> (1997). On nonparametric estimation of density level sets. <i>Ann. Statist.<\/i> <b>25<\/b> 948\u2013969.","DOI":"10.1214\/aos\/1069362732"},{"key":"34","doi-asserted-by":"publisher","unstructured":"<small>VAN DER VAART, A. W.<\/small> and <small>WELLNER, J. A.<\/small> (1996). <i>Weak Convergence and Empirical Processes. With Applications to Statistics<\/i>. <i>Springer Series in Statistics<\/i>. Springer, New York.","DOI":"10.1007\/978-1-4757-2545-2"},{"key":"35","doi-asserted-by":"publisher","unstructured":"<small>XU, M.<\/small> and <small>SAMWORTH, R. J.<\/small> (2021). High-dimensional nonparametric density estimation via symmetry and shape constraints. <i>Ann. Statist.<\/i> <b>49<\/b> 650\u2013672.","DOI":"10.1214\/20-aos1972"},{"key":"33","doi-asserted-by":"publisher","unstructured":"<small>van der Vaart, A. W.<\/small> (1998). <i>Asymptotic Statistics<\/i>. <i>Cambridge Series in Statistical and Probabilistic Mathematics<\/i> <b>3<\/b>. Cambridge Univ. Press, Cambridge.","DOI":"10.1017\/CBO9780511802256"},{"key":"3","doi-asserted-by":"publisher","unstructured":"<small>Brunel, V.-E.<\/small> (2018). Methods for estimation of convex sets. <i>Statist. Sci.<\/i> <b>33<\/b> 615\u2013632.","DOI":"10.1214\/18-STS669"},{"key":"24","doi-asserted-by":"publisher","unstructured":"<small>Polonik, W.<\/small> (1998). The silhouette, concentration functions and ML-density estimation under order restrictions. <i>Ann. Statist.<\/i> <b>26<\/b> 1857\u20131877.","DOI":"10.1214\/aos\/1024691360"},{"key":"12","doi-asserted-by":"publisher","unstructured":"<small>Groeneboom, P.<\/small> (1989). Brownian motion with a parabolic drift and Airy functions. <i>Probab. Theory Related Fields<\/i> <b>81<\/b> 79\u2013109.","DOI":"10.1007\/BF00343738"},{"key":"16","doi-asserted-by":"publisher","unstructured":"<small>Kim, J.<\/small> and <small>Pollard, D.<\/small> (1990). Cube root asymptotics. <i>Ann. Statist.<\/i> <b>18<\/b> 191\u2013219.","DOI":"10.1214\/aos\/1176347498"}],"container-title":["The Annals of Statistics"],"original-title":[],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T19:32:57Z","timestamp":1791401577000},"score":1,"resource":{"primary":{"URL":"https:\/\/projecteuclid.org\/journals\/annals-of-statistics\/volume-50\/issue-3\/Cube-root-weak-convergence-of-empirical-estimators-of-a-density\/10.1214\/21-AOS2157.full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,1]]},"references-count":35,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2022,6,1]]}},"URL":"https:\/\/doi.org\/10.1214\/21-aos2157","relation":{},"ISSN":["0090-5364","2168-8966"],"issn-type":[{"value":"0090-5364","type":"print"},{"value":"2168-8966","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,1]]}}}