Akaike Information Criterion for Selecting Components of the Mean Vector in High Dimensional Data with Fewer Observations
Akaike Information Criterion for Selecting Components of the Mean Vector in High Dimensional Data with Fewer Observations
复制标题
观测较少的高维数据中均值向量分量选择的 Akaike 信息准则
DOI：
--
复制
发表时间：
2008
期刊：
影响因子：
--
通讯作者：
T. Kubokawa
中科院分区：
文献类型：
--
作者：
M. Srivastava;T. Kubokawa
Akaike 信息准则 (AIC) 在文献中非常成功地用于少量参数和大量观测值 N 的模型选择。文献中尚未考虑 pi 大且接近 N 或 p>N 时的情况。事实上，当 pi 很大且接近 N 时，可用的 AIC 根本表现不佳。我们在寻找单样本多元分析中可能不为零的均值向量分量数量的背景下考虑这些情况。事实上，我们通过将其视为 Rao (1959) 以及 Potthoff 和 Roy (1964) 中引入的增长曲线模型来更普遍地考虑这个问题。通过仿真，结果表明所提出的 AIC 程序性能非常好。
The Akaike information criterion (AIC) has been used very successfully in the literature in model selection for small number of parameters pand large number of observations N. The cases when pis large and close to N or when p>N have not been considered in the literature. In fact, when pis large and close to N, the available AIC does not perform well at all. We consider these cases in the context of finding the number of components of the mean vector that may be different from zero in one-sample multivariate analysis. In fact, we consider this problem in more generality by considering it as a growth curve model introduced in Rao (1959) and Potthoff and Roy (1964). Using simulation, it has been shown that the proposed AIC procedures perform very well.