期刊信息

  • 刊名: 河北师范大学学报(自然科学版)Journal of Hebei Normal University (Natural Science)
  • 主办: 河北师范大学
  • ISSN: 1000-5854
  • CN: 13-1061/N
  • 中国科技核心期刊
  • 中国期刊方阵入选期刊
  • 中国高校优秀科技期刊
  • 华北优秀期刊
  • 河北省优秀科技期刊

线性混合模型中方差分量的非负估计

  • 1. 华北科技学院基础部部, 北京 101601;
    2. 石家庄职业技术学院基础部, 河北石家庄 050081
  • DOI:

Nonnegative Estimation of Variance Components in Linear Mixed Effects Models

摘要/Abstract

摘要:

在线性混合效应模型中,方差分量的非负估计问题备受关注.基于方差分量的ANOVA估计,给出了方差分量的不依赖于随机效应分布的非负改进,且改进后的估计在均方误差意义下优于ANOVA估计,有比ANO-VA估计更好的极限性质.当随机效应和随机误差都服从正态分布时,模拟显示,新估计优于ANOVA估计以及Tatsuya估计.

Abstract:

Nonneg ative estimator of variance components in linear mixed effects model are proposed based on their ANOVA estimator.These estimators have less mean squared errors than ANOVA estimator,and do not depend on the distributions of random effects.When random effects and errors are normal distribution,simulation shows that the new estimators are better than ANOVEA estimator and Tat suya estimator in the sense of meansquared error.