随机模型两种二次无偏估计方法的等价性

Equivalence of the two quadratic unbiased estimation methods for random model

分析最小范数二次无偏估计(M INQUE估计法)和最小方差二次无偏估计(B IQUE估计法),导出最小方差二次无偏估计的计算公式,证明这两种估计方法等价的条件.理论分析表明,无论初值如何选取,在迭代收敛的条件下,M INQUE估计法和B IQUE估计法的结果一致.最小方差二次无偏估计的意义明确,建模过程简明,易于理解.

The minimum norm quadratic unbiased estimation （MINQUE） and the minimum variance quadratic unbiased estimation （BIQUE） are analysed. The calculation formula to the minimum variance quadratic unbiased estimation is educed and the equivalent condition of the two estimations is proved. It is showed by the theoretical analysis, whatever the initial value is selected, it is conformable in the result of the minimum norm quadratic unbiased estimation and the minimum variance quadratic unbiased estimation in the iterative convergence condition. The meaning of the minimum variance quadratic unbiased estimation is explicit, and the modeling process is simple and easy to be comprehended.