Shrinkage estimation with a matrix loss function

Shrinkage estimation with a matrix loss function
复制标题

使用矩阵损失函数进行收缩估计

DOI:
10.1214/12-ejs748
复制
发表时间:
2011
期刊:
arXiv: Statistics Theory
影响因子:
--
通讯作者:
W. Strawderman
W. Strawderman
中科院分区:
--
文献类型:
--
作者:
R. Abu;J. Kent;W. Strawderman

文献摘要

参考文献

被引文献

相似文献

考虑估计具有恒定方差的独立正态分布观测值的n × p矩阵的n × p均值矩阵,其中估计器的性能是使用p × p矩阵二次误差损失函数来判断的。提出了一个矩阵版本的James-Stein估计量,依赖于一个调谐常数。当n大于或等于3时,对于某些调谐常数的选择,它优于通常的最大似然估计量。这一结果也适用于其他收缩估计器和设置。
Consider estimating the n by p matrix of means of an n by p matrix of independent normally distributed observations with constant variance, where the performance of an estimator is judged using a p by p matrix quadratic error loss function. A matrix version of the James-Stein estimator is proposed, depending on a tuning constant. It is shown to dominate the usual maximum likelihood estimator for some choices of of the tuning constant when n is greater than or equal to 3. This result also extends to other shrinkage estimators and settings.
协方差矩阵未知的正态均值矩阵的广义贝叶斯极小极大估计
DOI: --
发表时间: 2009
期刊: Journal of Multivariate Analysis 100
影响因子: --
作者:
Tsukuma;H.
通讯作者: H.