ESTIMATING THE PARAMETERS OF RARE EVENTS

被引：42

作者：

HSING, TL

机构：

[1] Department of Statistics, Texas A and M University, College Station

来源：

STOCHASTIC PROCESSES AND THEIR APPLICATIONS | 1991年 / 37卷 / 01期

基金：

美国国家科学基金会;

关键词：

EXTREMAL INDEX; ORDER STATISTIC; TAIL PROBABILITY;

D O I：

10.1016/0304-4149(91)90064-J

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：

In this paper the estimation of certain parameters of the extreme order statistics of stationary observations is considered in a general framework. These parameters are resulted from dependence, and hence their inference is substantially different from similar considerations in the i.i.d. context. Variants of estimators that are functionals of the empirical 'cluster size' distribution are proposed, and such properties as consistency and asymptotic normality are studied. Special emphasis is given to estimating the extremal index.

引用

页码：117 / 139

页数：23

共 20 条

[1]

[Anonymous], 1991, INTRO PROBABILITY TH

[2]

Billingsley P, 1968, CONVERGENCE PROBABIL

[3] TAIL ESTIMATES MOTIVATED BY EXTREME VALUE THEORY [J].