Communications in Statistics - Simulation and Computation.
ISSN/ISBN: Not available at this time. DOI: 10.1080/03610918.2022.2032153
Abstract: Hill pointed out in his landmark paper that “An interesting open problem is to determine which common distributions (or mixtures thereof) satisfy Benford’s law …”. Ever-since, there has been many attempts in finding distributions that are precisely compliant with Benford’s law. Even though sufficient conditions were derived and some ad-hoc distributions were reported in the literature, the lack of a general framework for generating such distributions is sensed. Almost all of the reported Benford-compliant distributions are finite-length. This paper looks at the problem from an electrical engineer’s perspective; it harnesses the literature on Nyquist inter-symbol interference theorem and then proposes a framework for generating infinite-length or arbitrary long finite-length distributions satisfying Benford’s law.
Bibtex:
@article{,
author = {Javad Kazemitabar},
title = {A general framework for constructing distributions satisfying Benford’s law},
journal = {Communications in Statistics - Simulation and Computation},
volume = {0},
number = {0},
pages = {1--8},
year = {2022},
publisher = {Taylor & Francis},
doi = {10.1080/03610918.2022.2032153},
URL = {https://doi.org/10.1080/03610918.2022.2032153
}
Reference Type: Journal Article
Subject Area(s): Probability Theory, Statistics