arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1403.6016math.STstat.TH
收藏

D-范数空间的重新审视

The space of D-norms revisited

Stefan Aulbach, Michael Falk, Maximilian Zott

摘要:

$D$ -范数理论是多元极值理论的产物。我们提出了最近的结果$D$ -范数,这是完全由一定的随机向量称为生成元。在第一部分中,我们证明了如果以适当的方式赋予Wasserstein-度量,则$D$ -范数空间是完全可分的度量空间。其次,将一个生成元与一个双随机阵相乘,得到另一个生成元。这个乘法的迭代提供了一个序列的$D$ -范数,我们计算其极限。最后,我们考虑一个参数族的$D$ -范数,我们假设生成元遵循对称Dirichlet分布。这个家庭涵盖了完全依赖和独立之间的整个范围。

英文摘要:

The theory of $D$-norms is an offspring of multivariate extreme value theory. We present recent results on $D$-norms, which are completely determined by a certain random vector called generator. In the first part it is shown that the space of $D$-norms is a complete separable metric space, if equipped with the Wasserstein-metric in a suitable way. Secondly, multiplying a generator with a doubly stochastic matrix yields another generator. An iteration of this multiplication provides a sequence of $D$-norms and we compute its limit. Finally, we consider a parametric family of $D$-norms, where we assume that the generator follows a symmetric Dirichlet distribution. This family covers the whole range between complete dependence and independence.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑