均值矩阵的线性估计在广义线性估计类中的可容许性  

THE ADMISSIBILITY OF LINEAR ESTIMATORS OF MEAN MATRIX IN THE CLASS OF GENERALIZED LINEAR ESTIMATORS

作　　者：杨镜华 

机构地区：[1]珠海广播电视大学,珠海519000

出　　处：《华南师范大学学报（自然科学版）》1999年第2期15-19,共5页Journal of South China Normal University(Natural Science Edition)

摘　　要：对于多元线性模型Ｙ～（ＸΘ，σ２ΣＶ），本文在最大特征值作为矩阵大小的比较标准下，讨论了ＸΘ的函数ＳＸΘ的线性估计ＡＹ在齐次线性估计类中可容许与ＡＹ在齐次广义线性估计类中可容许的关系，还讨论了ＡＹ＋Ｄ在广义线性估计类中可容许与ＡＹ在齐次广义线性估计类中可容许的关系－For the multivariate linear model Y～(XΘ,σ 2ΣV), where Σ>0 and V>0 are known, with the maximum eigenvalue as the criteria of comparison of matrix size, the relationship between the admissibility of AY in the class of homogenous linear estimators and in the class of homogenous generalized linear estimators is discussed. Also, the relationship between the admissibility of AY+D in the class of generalized linear estimators and the admissibility of AY in the class of homogenous generalized linear estimators is analyzed.

关 键 词：广义线性估计 均值矩阵 可容许性 线性估计 

分 类 号：O211.67[理学—概率论与数理统计] O212.4[理学—数学]