Concentration inequalities for high-dimensional linear processes with dependent innovations
Statistics Theory
2024-10-18 v2 Methodology
Machine Learning
Statistics Theory
Abstract
We develop concentration inequalities for the norm of vector linear processes with sub-Weibull, mixingale innovations. This inequality is used to obtain a concentration bound for the maximum entrywise norm of the lag- autocovariance matrix of linear processes. We apply these inequalities to sparse estimation of large-dimensional VAR(p) systems and heterocedasticity and autocorrelation consistent (HAC) high-dimensional covariance estimation.
Cite
@article{arxiv.2307.12395,
title = {Concentration inequalities for high-dimensional linear processes with dependent innovations},
author = {Eduardo Fonseca Mendes and Fellipe Lopes},
journal= {arXiv preprint arXiv:2307.12395},
year = {2024}
}