English

A Refined Probabilistic Error Bound for Sums

Numerical Analysis 2021-04-15 v1 Numerical Analysis

Abstract

This paper considers a probabilistic model for floating-point computation in which the roundoff errors are represented by bounded random variables with mean zero. Using this model, a probabilistic bound is derived for the forward error of the computed sum of n real numbers. This work improves upon existing probabilistic bounds by holding to all orders, and as a result provides informative bounds for larger problem sizes.

Keywords

Cite

@article{arxiv.2104.06531,
  title  = {A Refined Probabilistic Error Bound for Sums},
  author = {Eric Hallman},
  journal= {arXiv preprint arXiv:2104.06531},
  year   = {2021}
}
R2 v1 2026-06-24T01:08:31.965Z