摘要
在给定的线性模型下 ,讨论了在设计阵列亏秩时 ,最小二乘估计、最优加权最小二乘估计和线性无偏最小方差估计的性能比较 .得出了在一定条件下 ,最优加权最小二乘估计等价于线性无偏最小方差估计 .在噪声方差矩阵可逆、未知参数方差矩阵可逆条件下 ,可算出最优加权最小二乘估计与线性无偏最小方差估计方差的差表达式 ,并在一定条件下 。
The discussion on the performance coparison among least squares estimate, optimally weighted least squares estimate and linear unbiased minimum variance estimate of column singular design matrix for a linear model is presented. Under a condition on noise variance matrix invertibility and unknown parameter variance matrix invertilibity, the difference between optimally weighted least squares estimate and linear unbiased minimum variance estimate error variances can be calculated, and the two estimates can converge to the same one under a certain condition.
出处
《四川大学学报（自然科学版）》
CAS
CSCD
北大核心
2002年第5期839-842,共4页
Journal of Sichuan University(Natural Science Edition)
