Small deviations for fractional stable processes

被引:26
|
作者
Lifshits, M [1 ]
Simon, T [1 ]
机构
[1] Univ Evry Val Essonne, Equipe Anal & Prob, F-91025 Evry, France
来源
ANNALES DE L INSTITUT HENRI POINCARE-PROBABILITES ET STATISTIQUES | 2005年 / 41卷 / 04期
基金
俄罗斯基础研究基金会;
关键词
fractional Brownian motion; Gaussian process; linear fractional stable motion; Riemann-Liouville process; small ball constants; small ball probabilities; small deviations; stable process; wavelets;
D O I
10.1016/j.anihpb.2004.05.004
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Let {R-t, 0 <= t <= 1} be a symmetric alpha-stable Riemann-Liouville process with Hurst parameter H > 0. Consider a translation invariant, beta-self-similar, and p-pseudo-additive functional semi-norm vertical bar vertical bar (.) vertical bar vertical bar. We show that if H > beta + 1/p and gamma = (H - beta - 1/p)(-1), then lim epsilon(gamma)(epsilon down arrow 0) logP[vertical bar vertical bar R vertical bar vertical bar <= epsilon] = -K epsilon (-infinity, 0), with K finite in the Gaussian case alpha = 2. If alpha < 2, we prove that K is finite when R is continuous and H > beta + 1/p + 1/alpha. We also show that under the above assumptions, lim epsilon(gamma)(epsilon down arrow 0) logP[vertical bar vertical bar X vertical bar vertical bar <= epsilon] = -K epsilon (-infinity, 0), where X is the linear a-stable fractional motion with Hurst parameter H E (0, 1) (if a = 2, then X is the classical fractional Brownian motion). These general results cover many cases previously studied in the literature, and also prove the existence of new small deviation constants, both in Gaussian and non-Gaussian frameworks. (c) 2004 Elsevier SAS. All rights reserved.
引用
收藏
页码:725 / 752
页数:28
相关论文
共 50 条