中文摘要：

首先在X1，X2．…，Xn独立同分布（iidF）的情况下，给出了分位点过程n^1/2{Fn^-1（g）-F^-1（g）}分布的随机加权估计；其次证明了n^1/2。{Fn^-1（g）-F^-1（g）}的随机加权逼近的相合性；最后将随机加权估计应用于多传感器数据融合中。仿真结果表明：文中提出的随机加权估计优于文献[4]所给出的Bootstrap逼近，提高了估计的精度。

英文摘要：

The probability distribution of independent and identically distributed （i. i. d） random variables was estimated by random weighting method. The consistency of random weighting approximation on the distribution of n^1/2{Fn^-1（g）-F^-1（g）} was studied. Under the condition of random variables series being NA associated sample, sample mean＇s approximation problem was discussed. Negative associated sample between strongly stationary and independent classes was defined. The sample was divided into k classes and the weight 1/k was given to the k jackknifed virtual values Y; （i = 1 ,…, k） of Xn.Then the empirical distribution function can be obtained. The distribution of √n（Xn-μ） was simulated by the conditional distribution of √n（Yk^＊-Xn）, where independent sample Y1^＊,…,Yk^＊ was sampled from empirical distribution function Fk^＊. For the distribution of n1/2 {Fn^-1 （g） -F^-1 （g）}, the consistency of the random weighting approximation was proved.