常帥
(山西大學 數(shù)學科學學院,山西 太原 030006)
一類重尾二階參數(shù)的估計
常帥
(山西大學 數(shù)學科學學院,山西 太原 030006)
基于統(tǒng)計量M(α)n(k),文章提出了一類二階參數(shù)的估計,二階參數(shù)在極值理論中扮演著重要的角色,在重尾分布的尾指數(shù)估計中門限的最優(yōu)選取,以及尾指數(shù)的漸近偏差的降低都取決于二階參數(shù)ρ.然后,在極值理論的二階與三階正則條件下,討論了其漸近正態(tài)性.
二階參數(shù);正則條件;漸近正態(tài)性
[1] Celuk J,de Haan L.Regular Variation,Ectensions and Tauberian Theorems[M].CWI Tract 40,Center for Mathematics and Computer Science,Amsterdam,The Netherlands,1987.
[2] de Haan L,F(xiàn)erreira A.Extreme Value Theory:An Introduction[M].New York:Springer,2006.
[3] Gomes M I,de Haan L,Peng L.Semi-parametric Estimation of the Second Order Parameter in Statistics of Extremes[J].Extremes,2002,5(4):387-414.
A Class of Estimators of the Second Order Parameter for Heavy-tailed
CHANG Shuai
(SchoolofMathematicalSciences,ShanxiUniversity,Taiyuan030006,China)
Based on the statisticsM(α)n(k),we present a class of estimators of the second order parameter.The second order parameterρis of primordial importance in the adaptive choice of the best threshold to be considered in the estimation of the heavy-tailed index.Moreover,the classical extimators of the tail index try to reduce the main component of their asymptotic bias,which depends strongly onρ.Then,we study asymptotic normality of the second order parameter under the second order condition and the third order condition in the extreme value theory.
second order parameter;regular conditions;asymptotic normality
O212.1
A
0253-2395(2011)S2-0018-03
2011-09-03
常帥(1982-),男,山西大同人,碩士研究生,研究方向:時間序列分析.E-mail:307825111@163.com