不完全数据场合指数分布参数的置信区间

摘要

指数分布在可靠性统计中有着广泛的应用,对其总体参数进行估计具有重要的意义.本文针对一般的不完全数据场合,给出一种求解指数分布总体参数的置信区间的方法.该方法利用样本的中位数及相邻的次序统计量构造枢轴量,减小了不完全数据的影响.由于结果比较复杂,在样本量充分大时,利用枢轴量的渐进分布构造了总体参数近似的置信区间.%It is of great importance to estimate the parameter of exponential distribution because of its wide usefulness in reliability statistics.In this paper,a solution to the confidence interval for parameter of exponential distribution is discussed in the case of generally incomplete data.This method is given by constructing the pivot based on the sample median and its adjacent order statistic,which reduces the effect of incomplete data.Since the result is complicated,an approximate confidence interval for parameter is constructed by using the asymptotic distribution of the pivot when the sample size is large enough.