×

zbMATH — the first resource for mathematics

Adaptive density estimation using the blockwise Stein method. (English) Zbl 1098.62040
Summary: We study the problem of nonparametric estimation of a probability density of unknown smoothness in \(L_2(\mathbb R)\). Expressing the mean integrated squared error (MISE) in the Fourier domain, we show that it is close to the mean squared error in the Gaussian sequence model. Then applying a modified version of Stein’s blockwise method, we obtain a linear monotone oracle inequality.
Two consequences of this oracle inequality are that the proposed estimator is sharp minimax adaptive over a scale of Sobolev classes of densities, and that its MISE is asymptotically smaller than or equal to that of kernel density estimators with any bandwidth provided that the kernel belongs to a large class of functions including many standard kernels.
Reviewer: Reviewer (Berlin)

MSC:
62G07 Density estimation
PDF BibTeX XML Cite
Full Text: DOI Euclid
References:
[1] Birgé, L. and Massart, P. (2001) Gaussian model selection. J. Eur. Math. Soc., 3, 203-268. · Zbl 1037.62001 · doi:10.1007/s100970100031
[2] Boiko, L.L. and Golubev, G.K. (2000) How to improve the nonparametric density estimation in SPLUS. Problems Inform. Transmission, 36, 354-361. · Zbl 0971.62015
[3] Cai, T. (1999) Adaptive wavelet estimation: a block thresholding and oracle inequality approach. Ann. Statist., 27, 898-924. · Zbl 0954.62047 · doi:10.1214/aos/1018031262
[4] Cavalier, L. and Tsybakov, A.B. (2001) Penalized blockwise Steinś method, monotone oracles and sharp adaptive estimation. Math. Methods Statist., 10, 247-282. · Zbl 1005.62027
[5] Cavalier, L., Golubev, Yu., Picard, D. and Tsybakov, A.B. (2002) Oracle inequalities for inverse problems. Ann. Statist., 30, 843-874. · Zbl 1029.62032 · doi:10.1214/aos/1028674843 · euclid:aos/1028674843
[6] Cline, D.B.H. (1988) Admissible kernel estimators of a multivariate density. Ann. Statist., 16, 1421- 1427. · Zbl 0653.62027 · doi:10.1214/aos/1176351046
[7] Dalelane, C. (2005) Exact minimax risk for density estimators in non-integer Sobolev classes. Preprint no. 979, Laboratoire de Probabilités et Modèles Aléatoires, Universités Paris 6 and Paris 7 (http://www.proba.jussieu.fr/mathdoc/preprints). URL: · www.proba.jussieu.fr
[8] Devroye, L. and Penrod, C.S. (1984) Distribution-free lower bounds in density estimation. Ann. Statist., 12, 1250-1262. · Zbl 0551.62024 · doi:10.1214/aos/1176346790
[9] Donoho, D.L. and Johnstone, I.M. (1994) Ideal spatial adaptation via wavelet shrinkage. Biometrika, 81, 425-455. JSTOR: · Zbl 0815.62019 · doi:10.1093/biomet/81.3.425 · links.jstor.org
[10] Donoho, D.L. and Johnstone, I.M. (1995) Adapting to unknown smoothness via wavelet shrinkage. J. Amer. Statist. Assoc., 90, 1200-1224. JSTOR: · Zbl 0869.62024 · doi:10.2307/2291512 · links.jstor.org
[11] Efroimovich, S.Yu. (1985) Nonparametric estimation of a density with unknown smoothness. Theory Probab. Appl., 30, 557-568. · Zbl 0593.62034 · doi:10.1137/1130067
[12] Efroimovich, S.Yu. and Pinsker, M.S. (1982) Estimation of square-integrable probability density of a random variable Problems Inform. Transmission, 18, 175-189. · Zbl 0533.62038
[13] Efroimovich, S.Yu. and Pinsker, M.S. (1984) Learning algorithm for nonparametric filtering. Automat. Remote Control, 11, 1434-1440. · Zbl 0637.93069
[14] Efromovich, S.Yu. (1999) Nonparametric Curve Estimation. New York: Springer-Verlag. · Zbl 0935.62039
[15] Efromovich, S. (2000) On sharp adaptive estimation of multivariate curves. Math. Methods Statist., 9, 117-139. · Zbl 1006.62033
[16] Efromovich, S. (2004a) Oracle inequalities for Efromovich-Pinsker blockwise estimates. Methodol. Comput. Appl. Probab., 6, 303-322. · Zbl 1045.62020 · doi:10.1023/B:MCAP.0000026562.80429.48
[17] Efromovich, S. (2004b) Analysis of blockwise shrinkage wavelet estimates via lower bounds for nosignal setting. Ann. Inst. Statist. Math., 56, 205-223. · Zbl 1056.62048 · doi:10.1007/BF02530542
[18] Efromovich, S. (2005) Estimation of the density of regression errors. Ann. Statist., 33, 2194-2227. · Zbl 1086.62053 · doi:10.1214/009053605000000435
[19] Ekeland, I. and Temam, R. (1974) Analyse convexe et problèmes variationnels. Paris: Dunod Gauthier- Villars. · Zbl 0281.49001
[20] Goldenshluger, A., and Tsybakov, A. (2001) Adaptive prediction and estimation in linear regression with infinitely many parameters. Ann. Statist., 29, 1601-1619. · Zbl 1043.62076 · doi:10.1214/aos/1015345956
[21] Golubev, G.K. (1990) Quasi-linear estimates of signals in L2. Problems Inform. Transmission, 26, 15-20. · Zbl 0723.62026
[22] Golubev, G.K. (1991) LAN in nonparametric estimation of functions and lower bounds for quadratic risks. Theory Probab. Appl., 36, 152-157. · Zbl 0738.62043 · doi:10.1137/1136014
[23] Golubev, G.K. (1992) Nonparametric estimation of smooth probability densties in L2. Problems Inform. Transmission, 28, 44-54. · Zbl 0785.62039
[24] Golubev, G.K. and Levit B.Y. (1996) Distribution function estimation: adaptive smoothing. Math. Methods Statist., 5, 383-403. · Zbl 0902.62044
[25] Golubev, G.K. and Nussbaum, M. (1992) Adaptive spline estimates in a nonparametric regression model. Theory Probab. Appl., 37, 521-529. · Zbl 0787.62044 · doi:10.1137/1137102
[26] Hall, P., Kerkyacharian, G. and Picard, D. (1998) Block thresholding rules for curve estimation using kernel and wavelet methods. Ann. Statist., 26, 922-942. · Zbl 0929.62040 · doi:10.1214/aos/1024691082
[27] Härdle, W., Kerkyacharian, G., Picard, D., and Tsybakov, A. (1998) Wavelets, Approximation, and Statistical Applications, Lecture Notes in Statist. 129. New York: Springer-Verlag. · Zbl 0899.62002
[28] Kneip, A. (1994) Ordered linear smoothers. Ann. Statist., 22, 835-866. · Zbl 0815.62022 · doi:10.1214/aos/1176325498
[29] Li, K.-C. (1987) Asymptotic optimality of CP, CL, cross-validation and generalized cross-validation: Discrete index set. Ann. Statist., 15, 958-976. · Zbl 0653.62037 · doi:10.1214/aos/1176350486
[30] Lukacs, E. (1970). Characteristic Functions, 2nd edn. London: Griffin. · Zbl 0201.20404
[31] Lukacs, E. (1983) Developments in Characteristic Function Theory. London: Griffin. · Zbl 0515.60022
[32] Nemirovski, A. (2000) Topics in non-parametric statistics. In P. Bernard (ed.), Lectures on Probability Theory and Statistics. École d\'Été de Probabilités de Saint-Flour XXVIII - 1998, Lecture Notes in Math. 1738, pp. 85-277. Berlin: Springer-Verlag. · Zbl 0998.62033
[33] Polyak, B.T. and Tsybakov, A.B. (1992) A family of asymptotic optimal methods for choosing the estimate order in orthogonal series regression. Theory Probab. Appl., 37, 471-481. · Zbl 0806.62031
[34] Rigollet, R. (2004) Adaptive density estimation using Steinś blockwise method. Preprint no. 913, Laboratoire de Probabilités et Modèles Aléatoires, Universités Paris 6 and Paris 7 (http://www.proba.jussieu.fr/mathdoc/preprints). URL: · www.proba.jussieu.fr
[35] Schipper, M. (1996) Optimal rates and constants in L2-minimax estimation of probability density functions. Math. Methods Statist., 5, 253-274. · Zbl 0872.62043
[36] Shibata, R. (1981) An optimal selection of regression variables. Biometrika, 68, 45-54. JSTOR: · Zbl 0464.62054 · doi:10.1093/biomet/68.1.45 · links.jstor.org
[37] Stone, C.J. (1984) An asymptotically optimal window selection rule for kernel density estimates. Ann. Statist., 12, 1285-1297. · Zbl 0599.62052 · doi:10.1214/aos/1176346792
[38] Tsybakov, A. (2002) Discussion of ’Random rates in anisotropic regression\' by M Hoffmann and O Lepski. Ann. Statist., 30, 379-385.
[39] Tsybakov, A. (2004) Introduction à léstimation non-paramétrique. Berlin: Springer-Verlag. · Zbl 1029.62034
[40] Wegkamp, M.H. (1999) Quasi-universal bandwidth selection for kernel density estimators. Canad. J. Statist., 27, 409-420. JSTOR: · Zbl 1066.62521 · doi:10.2307/3315649 · links.jstor.org
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.