ASYMPTOTIC PROPERTIES OF GENERALIZED DPR STATISTIC

被引:0
|
作者
Vaiciulis, Marijus [1 ,2 ]
机构
[1] Vilnius State Univ, Inst Math & Informat, LT-08663 Vilnius, Lithuania
[2] Vilnius State Univ, Fac Math & Informat, LT-03225 Vilnius, Lithuania
关键词
asymptotic normality; DPR estimator; Hill estimator; Pickands estimator; mean-squared error; tail index; TAIL INDEX; ESTIMATORS; DISTRIBUTIONS; INFERENCE;
D O I
暂无
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
We investigate asymptotic properties of the so-called generalized Davydov, Paulauskas, and Rackauskas statistic that was introduced in [V. Paulauskas and M. Vaiciulis, Several modifications of DPR estimator of the tail index, Lith. Math. J., 51(1):36-50,2011]. We obtain an asymptotics of the bias of this statistic and show that this statistic is asymptotically normal. As an application, we provide a new class of semiparametric heavy-tailed estimators and compare these estimators with other estimators, including the popular Hill and Pickands estimators.
引用
收藏
页码:95 / 110
页数:16
相关论文
共 50 条