从有限样本到渐近:样条回归中准则选择的几何桥梁
统计理论
2007-06-13 v1 统计理论
摘要
本文在样条回归设定下研究选择准则的有限样本性质与其渐近对应物之间的联系,聚焦于弥合二者之间的差距。我们引入预测误差的偏差-方差分解,由此表明在渐近中偏差项主导变异性项,从而对该差距给出解释。为直观理解提供了几何阐释。理论与几何结果通过一个数值例子加以说明。
引用
@article{arxiv.math/0508596,
title = {From finite sample to asymptotics: A geometric bridge for selection criteria in spline regression},
author = {S. C. Kou},
journal= {arXiv preprint arXiv:math/0508596},
year = {2007}
}
备注
Published at http://dx.doi.org/10.1214/009053604000000841 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)