重尾分布高分位数的置信域
统计理论
2007-06-13 v1 统计理论
摘要
估计高分位数在风险管理中起着重要作用。这涉及对未知分布函数的外推。本文提出三种方法，即正态近似法、似然比法和数据倾斜法，来构造重尾分布高分位数的置信域。一项模拟研究倾向于数据倾斜法。
引用
@article{arxiv.math/0611278,
title = {Confidence regions for high quantiles of a heavy tailed distribution},
author = {Liang Peng and Yongcheng Qi},
journal= {arXiv preprint arXiv:math/0611278},
year = {2007}
}
备注
Published at http://dx.doi.org/10.1214/009053606000000416 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)