线性混合模型中方差分量的广义推断

本文考虑了线性混合模型中方差分量的假设检验和区间估计问题.基于广义P-值和广义置信区间的概念,构造了对应于随机效应的单个方差分量的精确检验和置信区间.所构造的广义p-值和广义置信区间是最小充分统计量的函数.对于两个独立线性混合模型中对应于随机效应的方差分量的比较,建立了精确检验和置信区间.进-步,研究了所给检验和置信区间的统计性质,给出了这些检验方法与文献中已有方法的功效比较的模拟结果.模拟结果表明,新检验在功效方面有显著的改进.最后,通过-个实例来演示本文方怯.

Generalized Inferences on the Variance Components in General Linear Mixed Model

In this paper, we consider the problems of hypothesis testing and interval estimation for variance components in general linear mixed model. Exact test and confidence interval for a single variance component corresponding to random effect are developed based on the concepts of generalized p-value and generalized confidence interval. The generalized p-values and generalized confidence intervals we have developed are functions of the minimal sufficient statistics. Exact test and confidence interval are also established for comparing the random-effects variance components in two independent general linear mixed models. Furthermore, we investigate the statistical properties of the resulting tests and confidence intervals. Some simulation results to compare the powers of the proposed test with those of the existing method are reported. The simulation results indicate that new test appears to have significant gain in the power. Finally, the proposed methods are demonstrated by a real example.