Back to Search Start Over

Berry-Esséen bounds and almost sure CLT for the quadratic variation of a class of Gaussian process.

Authors :
Chen, Yong
Ding, Zhen
Li, Ying
Source :
Communications in Statistics: Theory & Methods. 2024, Vol. 53 Issue 11, p3920-3939. 20p.
Publication Year :
2024

Abstract

We propose a condition which is valid for a class of continuous Gaussian processes that may fail to be self-similar or have stationary increments. Some examples include the sub-fractional Brownian motion and the bi-fractional Brownian motion and the sub-bifractional Brownian motion. Under this assumption, we show an upper bound for the difference between the inner product associated with a class of Gaussian process and that associated with the fractional Brownian motion. This inequality relates a class of Gaussian processes to the well studied fractional Brownian motion, which characterizes their relationship quantitatively. As an application, we obtain the optimal Berry-Esséen bounds for the quadratic variation when H ∈ (0 , 2 3 ] and the upper Berry-Esséen bounds when H ∈ (2 3 , 3 4 ]. As a by-product, we also show the almost sure central limit theorem (ASCLT) for the quadratic variation when H ∈ (0 , 3 4 ]. The results in the present paper extend and improve those in the literature. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
03610926
Volume :
53
Issue :
11
Database :
Academic Search Index
Journal :
Communications in Statistics: Theory & Methods
Publication Type :
Academic Journal
Accession number :
176582855
Full Text :
https://doi.org/10.1080/03610926.2023.2167055