Variance components in errors-in-variables models: estimability, stability and bias analysis

Variance components in errors-in-variables models: estimability, stability and bias analysis
复制标题

变量误差模型中的方差分量:可估计性、稳定性和偏差分析

DOI:
10.1007/s00190-014-0717-9
复制
发表时间:
2014-04
期刊:
影响因子:
4.4
通讯作者:
Liu Jingnan
Liu Jingnan
中科院分区:
地球科学1区
文献类型:
--
作者:
Xu Peiliang;Liu Jingnan

文献摘要

参考文献

被引文献

相似文献

尽管总体最小二乘在理论和实际应用中已经得到了大量的研究,但几乎没有同时解决参数估计和变量误差随机模型的问题。我们证明了EIV随机模型的方差成分是不可估计的,如果随机系数矩阵的元素可以分为两组或多组相同精度的数据。这种不可估计性的结果是令人惊讶的,因为它表明我们没有办法获得关于这种EIV随机模型的任何知识。我们证明了如果方差分量在理论上是可估计的,那么估计方差分量的线性方程可能是病态的。最后,如果方差成分是可估计的,我们推导出它们估计的偏差,由于条件数很大,这种偏差可能会被显著放大。
Although total least squares has been substantially investigated theoretically and widely applied in practical applications, almost nothing has been done to simultaneously address the estimation of parameters and the errors-in-variables (EIV) stochastic model. We prove that the variance components of the EIV stochastic model are not estimable, if the elements of the random coefficient matrix can be classified into two or more groups of data of the same accuracy. This result of inestimability is surprising as it indicates that we have no way of gaining any knowledge on such an EIV stochastic model. We demonstrate that the linear equations for the estimation of variance components could be ill-conditioned, if the variance components are theoretically estimable. Finally, if the variance components are estimable, we derive the biases of their estimates, which could be significantly amplified due to a large condition number.
DOI: 10.2307/2348295
发表时间: 1991
期刊: The Statistician
影响因子: --
作者:
R. Iyer;J. Magnus;H. Neudecker
通讯作者: R. Iyer;J. Magnus;H. Neudecker
DOI: 10.2307/2531661
发表时间: 1990-03
期刊: Biometrics
影响因子: 1.9
作者:
W. Venables;C. Rao;J. Kleffe
通讯作者: W. Venables;C. Rao;J. Kleffe
DOI: 10.1007/bf02522340
发表时间: 1986-12
期刊: Bulletin Gæodésique
影响因子: --
作者:
K. Koch
通讯作者: K. Koch
DOI: 10.1201/9781315367002-6
发表时间: 2008
期刊: --
影响因子: --
作者:
Jonathan May
通讯作者: Jonathan May
DOI: 10.1080/02331887408801162
发表时间: 1974
期刊: Statistics
影响因子: 1.9
作者:
R. Pincus
通讯作者: R. Pincus