IMPROVED MINIMAX ESTIMATION OF POWERS OF THE VARIANCE OF A MULTIVARIATE NORMAL-DISTRIBUTION UNDER THE ENTROPY LOSS FUNCTION

被引:7
|
作者
PAL, N [1 ]
LING, CH [1 ]
机构
[1] UNIV SW LOUISIANA,DEPT STAT,LAFAYETTE,LA 70504
关键词
AFFINE EQUIVARIANCE; LOSS FUNCTION; MINIMAX ESTIMATOR; RISK FUNCTION; UNIFORM DOMINATION;
D O I
10.1016/0167-7152(94)00172-5
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
Consider the problem of estimating positive powers of the variance of a multivariate normal distribution. We show that Strawderman (Strawderman, 1974) type minimax estimators, which are better than the best affine equivariant estimators under the quadratic loss function, can also be derived under the entropy loss function.
引用
收藏
页码:205 / 211
页数:7
相关论文
共 50 条