The asymptotic minimax constant for sup-norm loss in nonparametric density estimation

被引:16
作者
Korostelev, A [1 ]
Nussbaum, M
机构
[1] Wayne State Univ, Dept Math, Detroit, MI 48202 USA
[2] Karl Weierstrass Inst Math, D-10117 Berlin, Germany
关键词
density estimation; exact constant; optimal recovery; uniform norm risk; white noise;
D O I
10.2307/3318561
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
We develop the exact constant of the risk asymptotics in the uniform norm for density estimation. This constant has already been found for nonparametric regression and for signal estimation in Gaussian white noise. Holder classes for arbitrary smoothness index beta>0 on the unit interval are considered. The constant involves the value of an optimal recovery problem as in the white noise case, but in addition it depends on the maximum of densities in the function class.
引用
收藏
页码:1099 / 1118
页数:20
相关论文
共 17 条
[1]  
Besov O.V., 1979, INTEGRAL REPRESENTAT, VII
[2]  
Brown LD, 1998, ANN STAT, V26, P279
[3]   ASYMPTOTIC MINIMAX RISK FOR SUP-NORM LOSS - SOLUTION VIA OPTIMAL RECOVERY [J].
DONOHO, DL .
PROBABILITY THEORY AND RELATED FIELDS, 1994, 99 (02) :145-170
[4]  
Freidlin M. I., 1984, RANDOM PERTURBATION
[5]  
Gyorfi L., 1985, The L 1 View
[6]  
IBRAGIMOV IA, 1982, THEOR PROBAB APPL, V27, P84
[7]  
KHASMINSKII RZ, 1978, THEOR PROBAB APPL+, V23, P794
[8]  
KOROSTELEV A, 1995, 154 WEIERSTR I
[9]  
KOROSTELEV AP, 1993, THEOR PROBAB APPL, V38, P775
[10]  
LEONOV SL, 1997, MATH METHODS STAT, V6, P476