Assessing the omission of records from a data set using Benford’s law
| Pages | 798-805 |
| Published date | 03 October 2016 |
| DOI | https://doi.org/10.1108/JFC-10-2015-0060 |
| Date | 03 October 2016 |
| Author | Pedro Carreira,Carlos Gomes da Silva |
| Subject Matter | Accounting & Finance,Financial risk/company failure,Financial crime |
Assessing the omission of
records from a data set using
Benford’s law
Pedro Carreira
School of Technology and Management and CIGS Management for
Sustainability Research Center, Leiria Polytechnic Institute, Leiria, Portugal, and
Carlos Gomes da Silva
School of Technology and Management, Leiria Polytechnic Institute,
Leiria, Portugal and INESC Coimbra, Coimbra, Portugal
Abstract
Purpose – The purpose of this paper is to propose a methodology to estimate the number of records
that were omitted from a data set, and to assess its effectiveness.
Design/methodology/approach – The procedure to estimate the number of records that were
omitted from a data set is based on Benford’s law. Empirical experiments are performed to illustrate the
application of the procedure. In detail, two simulated Benford-conforming data sets are distorted and
the procedure is then used to recover the original patterns of the data sets.
Findings – The effectiveness of the procedure seems to increase with the degree of conformity of the
original data set with Benford’s law.
Practical implications – This work can be useful in auditing and economic crime detection, namely
in identifying tax evasion.
Originality/value – This work is the rst to propose Benford’s law as a tool to detect data evasion.
Keywords Financial crime, Auditing, Benford’s law, Digital analysis, Fraud detection
Paper type Research paper
1. Introduction
The decisions made by economic agents are based on the disclosed and exchanged
information between them. By omitting some information from a data set (as, for
example, when some sales are unrecorded by rms), economic agents harm the
functioning of the economy. In the scal domain, evasion practices may lead to
decreased government revenue that could be used to raise global welfare. Also, at a
nancial level, biased information may lead economic agents to invest more frequently
in undesirable projects or to neglect protable ones.
The development of tools to audit and monitor the authenticity of high ows of
numerical information, especially those supported by information technologies, is thus
of relevance. Digital analysis, dened as the study of digits and patterns of numbers
(Nigrini and Mittermaier, 1997), has proved to be useful with this respect, as its
underlying techniques (which can be found for example in Coderre, 2009) can easily deal
with huge amounts of data, while being simple and automated.
A particular digital analysis technique consists of the application of Benford’s law
(Newcomb, 1881;Benford, 1938), which can be used to detect misbehavior of samples of
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1359-0790.htm
JFC
23,4
798
Journalof Financial Crime
Vol.23 No. 4, 2016
pp.798-805
©Emerald Group Publishing Limited
1359-0790
DOI 10.1108/JFC-10-2015-0060
To continue reading
Request your trial