Path developments and tail asymptotics of signature for pure rough paths

Path developments and tail asymptotics of signature for pure rough paths
复制标题

纯粗糙路径的路径发展和签名尾渐近

DOI:
10.1016/j.aim.2020.107043
复制
发表时间:
2020
影响因子:
1.7
通讯作者:
Boedihardjo H
Boedihardjo H
中科院分区:
数学1区
文献类型:
--
作者:
Boedihardjo H

文献摘要

参考文献

被引文献

相似文献

线性控制微分方程的解可以表示为沿驱动路径沿着的全局迭代路径积分。这种迭代积分的集合基本上编码了关于底层路径的所有信息。虽然迭代路径积分的上界是众所周知的,但下界知之甚少,并且仅相对最近才知道可以使用n阶迭代积分的某些类型的渐近性来恢复路径的某些内在定量性质,例如C1路径的长度。在本文中,我们研究了最简单的粗糙路径类型,(直线段的粗糙路径模拟),并根据基本路径的局部变差建立了迭代积分的尾渐近性的一致上下估计.我们的方法,我们认为这是这个问题的新方法,涉及到将路径发展成复半单李代数,并利用相关表示理论研究李代数发展下李多项式的谱性质。
Solutions to linear controlled differential equations can be expressed in terms of global iterated path integrals along the driving path. This collection of iterated integrals encodes essentially all information about the underlying path. While upper bounds for iterated path integrals are well known, lower bounds are much less understood, and it is known only relatively recently that some types of asymptotics for the n-th order iterated integral can be used to recover some intrinsic quantitative properties of the path, such as the length for C 1 paths. In the present paper, we investigate the simplest type of rough paths (the rough path analogue of line segments), and establish uniform upper and lower estimates for the tail asymptotics of iterated integrals in terms of the local variation of the underlying path. Our methodology, which we believe is new for this problem, involves developing paths into complex semisimple Lie algebras and using the associated representation theory to study spectral properties of Lie polynomials under the Lie algebraic development.
DOI: --
发表时间: 2015
期刊:
影响因子: --
作者:
Terry Lyons;Weijun Xu
通讯作者: Weijun Xu
DOI: --
发表时间: 2005
期刊:
影响因子: --
作者:
T. Kobayashi;T. Oshima
通讯作者: T. Oshima
DOI: --
发表时间: 2013
期刊:
影响因子: --
作者:
I. Chevyrev
通讯作者: I. Chevyrev
DOI: --
发表时间: 1986
期刊:
影响因子: --
作者:
N. N. Vorob’ev
通讯作者: N. N. Vorob’ev
线性粗微分方程的可积性(以及高斯过程迭代积分的矩估计)
DOI: --
发表时间: 2011
期刊:
影响因子: --
作者:
P. Friz;Sebastian Riedel
通讯作者: Sebastian Riedel