On Berry--Esseen bounds for non-instantaneous filters of linear processes

TL;DR

This paper derives Berry--Esseen bounds for sums of non-instantaneous functions of linear processes with long-range dependence, extending CLT results to dependent, non-linear, and long-memory contexts.

Contribution

It develops a finite orthogonal expansion approach to establish Berry--Esseen bounds for non-linear functionals of long-memory linear processes.

Findings

Established Berry--Esseen bounds for sums of non-linear functions of long-memory processes.

Extended CLT applicability to dependent, non-linear, long-range dependent sequences.

Provided bounds for a class of functions including indicators and polynomials.

Abstract

Let $X_n=\sum_{i=1}^{\infty}a_i\epsilon_{n-i}$, where the $\epsilon_i$ are i.i.d. with mean 0 and at least finite second moment, and the $a_i$ are assumed to satisfy $|a_i|=O(i^{-\beta})$ with $\beta >1/2$. When $1/2<\beta<1$, $X_n$ is usually called a long-range dependent or long-memory process. For a certain class of Borel functions $K(x_1,...,x_{d+1})$, $d\ge0$, from ${\mathcal{R}}^{d+1}$ to $\mathcal{R}$, which includes indicator functions and polynomials, the stationary sequence $K(X_n,X_{n+1},...,X_{n+d})$ is considered. By developing a finite orthogonal expansion of $K(X_n,...,X_{n+d})$, the Berry--Esseen type bounds for the normalized sum $Q_N/\sqrt{N},Q_N=\sum_{n=1}^N(K(X_ n,...,X_{n+d})-\mathrm{E}K(X_n,...,X_{n+d}))$ are obtained when $Q_N/\sqrt{N}$ obeys the central limit theorem with positive limiting variance.