多个逆高斯总体共同均值的广义统计推断

摘要

This paper presents procedures for hypothesis testing and interval estimation for the common mean of several inverse Gaussian populations when the scalar parameters are unknown and unequal.The proposed approach is the hybrid of the generalized inference method and the large-sample theory.Some simulation results are presented to compare the performances of the proposed approach with those of the existing approach.The simulation results indicate that the proposed approach has larger coverage probabilities than the existing approach.Also it has shorter expected lengths and higher powers than the existing approach when the scalar parameters are obviously different from each other.%研究了多个尺度参数未知且不等的逆高斯总体的共同均值的假设检验和区间估计问题,结合了广义推断理论和大样本理论,通过数值模拟比较了本文方法和已有方法。模拟结果表明,本文方法与已有方法相比具有较好的区间覆盖率,且当尺度参数相差较大时具有较短的区间长度和较大的功效。