Annals of Faculty of Economics, University of Oradea, Faculty of Economics, vol. 1(2), pp. 229-238.
ISSN/ISBN: Not available at this time. DOI: Not available at this time.
Abstract: The frequency of accounting data frauds has been increased in corporate environment. As a result of that, the research on detection of such irregularities in accounting and auditing is gaining researchers’ focus. Bedford’s law has been in the literature for the identification of data manipulation in accounting and auditing field. The application of this law in accounting fraud detection started in 1988 after the work of Carslaw (he observed a greater frequency of zeros and less frequency of nines in the second place in the reported earning numbers). The underlying idea about this technique is based on comparison of certain digit frequency to the expected digit pattern proposed by Bedford’s law. Various goodness-of-fit test are used to analyze the data conformity to Benford’s law based on a null hypothesis of conformity of empirical data to expected data pattern. This study addresses some of the most important goodness-of fit tests that can be used to analyze data pattern and digit behavior. Most importantly chi-square, Kolmogorov-Smirnov test (KS), Euclidean distance, Joenssen’s JP-square, Freedman-Watson u-square, Chebyshev distance, Z-statistics and mean absolute deviation tests are discussed with expression to calculate test statistics. Tests like Chi-Square and KS are also sensitive to size of data set, so a combination of various goodness-of fit test is recommended in literature to make more accurate analysis of data conformity.
Bibtex:
@Article{,
author={Rabeea Sadaf},
title={{Advanced Statistical Techniques For Testing Benford'S Law}},
journal={Annals of Faculty of Economics},
year=2017,
volume={1},
number={2},
pages={229--238},
month={December},
doi={},
url={https://ideas.repec.org/a/ora/journl/v1y2017i2p229-238.html}
}
Reference Type: Journal Article
Subject Area(s): Economics, Statistics