ON SMALL SAMPLE PROPERTIES OF THE ALMOST UNBIASED GENERALIZED RIDGE ESTIMATOR
ON SMALL SAMPLE PROPERTIES OF THE ALMOST UNBIASED GENERALIZED RIDGE ESTIMATOR
复制标题
DOI：
10.1080/03610928608829203
复制
发表时间：
1986-01-01
影响因子：
0.8
通讯作者：
OHTANI, K
中科院分区：
文献类型：
--
作者：
OHTANI, K
本文研究了运算几乎无偏广义岭估计（E）的小样本性质。导出了theAUGRE的精确前两阶矩。结果表明，虽然AUGRE的偏差有很大的减小，但在较大的非中心性参数范围内，AUGRE的均方误差比不作偏差校正的广义岭估计的效率要低得多.
The purpose of this paper is to examine small sample properties of the operational almost unbiased generalized ridge estimator (E) . The exact first two moments of theAUGRE are derived. It is shown that although the reduction of the bias of the AUGRE is substantial, the AUGRE is rather inefficient than the generalized ridge estimator without the bias correction in a wide range of a noncen-trality parameter in terms of the mean square error.