Approximation of Sums of Locally Dependent Random Variables via Perturbation of Stein Operator

TL;DR

This paper develops a Stein's method-based approach to approximate sums of locally dependent nonnegative integer-valued variables with Poisson, binomial, or negative binomial distributions, providing improved error bounds.

Contribution

It introduces a general error bound for total variation distance in approximating sums of dependent variables, with applications to runs, achieving better convergence rates than previous results.

Findings

Achieves $O(|J|^{-1})$ error bounds for runs.

Improves previous bounds from $O(|J|^{-0.5})$ and $O(1)$.

Provides a unified Stein's method framework for distribution approximation.

Abstract

Let $(X_{i}, i\in J)$ be a family of locally dependent nonnegative integer-valued random variables, and consider the sum $W=\sum\nolimits_{i\in J}X_i$. We first establish a general error upper bound for $d_{TV}(W, M)$ using Stein's method, where the target variable $M$ is either the mixture of Poisson distribution and binomial or negative binomial distribution. As applications, we attain $O(|J|^{-1})$ error bounds for ($k_{1},k_{2}$)-runs and $k$-runs under some special cases. Our results are significant improvements of the existing results in literature, say $O(|J|^{-0.5})$ in Pek\"{o}z [Bernoulli, 19 (2013)] and $O(1)$ in Upadhye, et al. [Bernoulli, 23 (2017)].