摘要
目的探讨两两NQD随机变量序列的密度核估计是否具有与NA序列密度核估计类似的相合性.方法设{Xn,n 1}为同分布的两两NQD随机变量序列,f(x)为X1的概率密度函数.基于样本X1,X2,…,Xn,给出了密度函数f(x)的核估计,在一定条件下,结合NA序列的相关结果的证明方法,引证后经过认真严谨的推导得出结论.结果(1)两两NQD序列的r阶平均相合性:(i)limn→∞E|fn(x)-f(x)|r=0,(ii)E|fn(x)-f(x)|r=O(n-r4);(2)逐点强相合性:fn(x)-f(x)→0,a.s.,对f(x)的任何连续点x成立;(3)一致强相合性:limn→∞sxu∈Ip|fn(x)-f(x)|=0,a.s.结论两两NQD随机变量序列的密度核估计具有较好的相合性.
Objective To discuss the consistencies of the kernel-type density estimation in the case of pariwise NQD random sequences. Method Let {Xn, n≥1} be a sequence of identically distributed and pariwise NQD random variables with probability density function f(x). Based on pairwise NQD samples, the kernel estimator for f (x) is constructed. Under some conditions, referring to the result of NA random sequences, this paper derives the result seriously and rigorously. Result (1) The consistency in rorder mean: (i) limE | fn(x) --f(x)|r=0, (ii) E|f(x) --f(x) | r=O (nr/4); (2) the pointwise strong consistency: fn (x) --f (x) →0, a. s. (3) the strong uniform consistency is shown under suitable conditions: limsup|fn(x)-f(x)|=0, a.s.. Conclusion The kernel-type density estimation in the case of pariwise NQD random sequences possesses the consistencies.
出处
《河北北方学院学报(自然科学版)》
2009年第4期1-6,共6页
Journal of Hebei North University:Natural Science Edition
关键词
两两NQD序列
NQD序列
密度函数核估计
相合性
pariwise NQD random variables sequences
negatively associated random variables sequences
the kernel estimator for density function
consistency