IMA Journal of Numerical Analysis 2(3), 249-274.
ISSN/ISBN: 0272-4979 DOI: Not available at this time.
Abstract: ABSTRACT: Extensions are made of a recently developed theory of floating-point error analysis to facilitate the construction and computation of strict error bounds of a posteriori type. Inter alia, these extensions obviate the need for specially-directed rounding procedures in computer hardware or software, and reduce computational cost by enabling error bounds to be computed to a low working precision. Applications are made to the evaluation of products, quotients, powers, sums and inner products, and also to the processes of input and output
Not available at this time.
Reference Type: Journal Article
Subject Area(s): Numerical Analysis