Superefficiency of a projection density estimator

被引:2
作者
Bosq, D [1 ]
机构
[1] Univ Paris 06, Lab Stat Theor & Appl, FR-75252 Paris, France
来源
CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE | 2005年 / 33卷 / 01期
关键词
adaptive estimation; data-driven truncation index; projection density estimator; superoptimal rates;
D O I
10.1002/cjs.5540330103
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
The author constructs a projection density estimator with a data-driven truncation index. This estimator reaches the superoptimal rates 1/n in mean integrated square error and {ln ln(n)/n}(1/2) in uniform almost sure convergence over a given subspace which is dense in the class of all possible densities; the rate of the estimator is quasi-optimal everywhere else. The subspace in question may be chosen a priori by the statistician.
引用
收藏
页码:21 / 37
页数:17
相关论文
共 23 条
[1]  
[Anonymous], SANKHYA A
[2]  
Billingsley P, 1968, CONVERGE PROBAB MEAS
[3]  
Bleuez J., 1979, REV ROUMAINE MATH PU, V24, P869
[4]   Locally superoptimal and adaptive projection density estimators [J].
Bosq, D .
COMPTES RENDUS MATHEMATIQUE, 2002, 334 (07) :591-595
[5]  
BOSQ D, 2002, ESTIMATION SUROPTIMA
[6]  
Bosq D., 1987, THEORIE ESTIMATION F
[7]  
Cencov N. N., 1962, SOV MATH, V3, P1559
[8]  
DEOLIVEIRA JT, 1963, REV FACULDADE CIENCI, V9, P213
[9]  
Devroye L., 1987, A course in density estimation
[10]  
Donoho DL, 1996, ANN STAT, V24, P508