Polynomial birth-death distribution approximation in Wasserstein distance

TL;DR

This paper provides probabilistic proofs for Stein's factors in polynomial birth-death distribution approximation within Wasserstein distance, extending previous work and demonstrating superior accuracy over Poisson-based approximations.

Contribution

It offers new probabilistic proofs for Stein's factors for PBD approximation and generalizes prior Poisson approximation results in Wasserstein distance.

Findings

Established upper bounds for Wasserstein distance between PBD and Poisson binomial distributions.

Showed PBD approximation is more accurate than Poisson or shifted Poisson.

Extended previous work by Brown & Xia (2001) and Barbour & Xia (2006).

Abstract

The polynomial birth-death distribution (abbr. as PBD) on $\ci=\{0,1,2, >...\}$ or $\ci=\{0,1,2, ..., m\}$ for some finite $m$ introduced in Brown & Xia (2001) is the equilibrium distribution of the birth-death process with birth rates $\{\alpha_i\}$ and death rates $\{\beta_i\}$, where $\a_i\ge0$ and $\b_i\ge0$ are polynomial functions of $i\in\ci$. The family includes Poisson, negative binomial, binomial and hypergeometric distributions. In this paper, we give probabilistic proofs of various Stein's factors for the PBD approximation with $\a_i=a$ and $\b_i=i+bi(i-1)$ in terms of the Wasserstein distance. The paper complements the work of Brown & Xia (2001) and generalizes the work of Barbour & Xia (2006) where Poisson approximation ($b=0$) in the Wasserstein distance is investigated. As an application, we establish an upper bound for the Wasserstein distance between the PBD and Poisson binomial distribution and show that the PBD approximation to the Poisson binomial distribution is much more precise than the approximation by the Poisson or shifted Poisson distributions.