摘要
{Xn,n≥1}是独立同分布随机变量序列,M(n1),M(n2)分别表示{X1,X2,…,Xn}的第一个最大值与第二个最大值.存在an>0,bn使得P(M(n1)≤anx+bn)→wG(x)成立(其中G(x)为极值指数分布),则对x>y有limN→∞1/logN sum from n=1 to N (1/nI{M_n^((1))≤u_n,M_n^((2))≤vn=G(y){log G(x)-log G(y)+1} a.s.)其中un=anx+bn,vn=any+bn.
Let ( Xn) be a sequence of i. i. d. random variables with distribution function F(x), Mn^(1) , Mn^(2) denote, respectively, the first and the second largest maximum of {X1 , X2, …, Xn }, assume also that there are normalizing sequences an〉0, bn and a nodegenerate limit distributionG(X),such that P(Mn^(1)≤anx+bn)→wG(x),then for x〉y we have an almost sure central limit theorem for Mn^(1) and Mn^(2),i,e.lim N→∞1/logN∑Nn=1 1/nI{Mn^(1)≤Mn^(2)≤vn}=G(y){logG(x)-log G(y)+1}a.s. where un=anx+bn,vn=any+bn.
出处
《西南大学学报(自然科学版)》
CAS
CSCD
北大核心
2008年第9期20-24,共5页
Journal of Southwest University(Natural Science Edition)
基金
国家自然科学基金资助项目(70371061)
重庆市自然科学基金资助项目(CSTC,2005BB8098)
关键词
几乎处处中心极限定理
非退化分布
极端顺序统计量
almost sure central limit theorem
nondegenerate limit distribution
extreme order statistics