Estimation of canonical dependence parameters a class of bivariate peaks-over-threshold models

被引：6

作者：

Falk, M

Reiss, RD

机构：

[1] Univ Siegen, FB 6, D-57068 Siegen, Germany

[2] Katholische Univ Eichstatt, Math Geog Fak, D-85071 Eichstatt, Germany

来源：

STATISTICS & PROBABILITY LETTERS | 2001年 / 52卷 / 03期

关键词：

generalized pareto distributions; dependence parameters; asymptotic normality; best linear unbiased estimator (BLUE);

D O I：

10.1016/S0167-7152(00)00194-2

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：

This paper deals with the estimation of dependence parameters in certain bivariate generalized Pareto models which are models for exceedances (peaks) over high thresholds. A unified approach is obtained by using canonical parameters. An estimator, which is related to a best linear unbiased estimator, turns out to be inefficient compared to a nonlinear one. (C) 2001 Elsevier Science B.V. All rights reserved.

引用

页码：233 / 242

页数：10

共 13 条

[1]

FALK M, 1994, LAWS SMALL NUMBERS E, V23

[2]

Galambos J., 1978, The asymptotic theory of extreme order statistics

[3]

Gumbel E. J., 1958, Statistics of Extremes

[4] MAXIMA OF NORMAL RANDOM VECTORS - BETWEEN INDEPENDENCE AND COMPLETE DEPENDENCE [J].