OPTIMAL CHOICE OF SAMPLE FRACTION IN EXTREME-VALUE ESTIMATION

被引:76
|
作者
DEKKERS, ALM [1 ]
DEHAAN, L [1 ]
机构
[1] ERASMUS UNIV ROTTERDAM,3000 DR ROTTERDAM,NETHERLANDS
关键词
EXTREME-VALUE THEORY; ORDER STATISTICS; ASYMPTOTIC NORMALITY; MEAN SQUARED ERROR; REGULAR VARIATION; PI-VARIATION; INVERSE COMPLEMENTARY FUNCTION;
D O I
10.1006/jmva.1993.1078
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 study the asymptotic bias of the moment estimator γ̂n for the extreme-value index γ ∈ R under quite natural and general conditions on the underlying distribution function. Furthermore the optimal choice for the sample franction in estimating γ is considered by minimizing the mean squared error of γ̂n - γ. The results cover all three limiting types of extreme-value theory. The connection between statistics and regular variation and Π-variation is handled in a systematic way. © 1993 Academic Press, Inc.
引用
收藏
页码:173 / 195
页数:23
相关论文
共 50 条