Detecting Digital Frequency Anomalies as Benchmarked against the Newcomb-Benford Theoretical Frequencies: Calibrating the Chi2 Test: A Note


  •  Edward Lusk    
  •  Michael Halperin    

Abstract

Digital Frequency Testing (DFT) using the Newcomb-Benford (N-B) profile is a simple and potentially effective tool that can be used efficiently in the certification audit as well as in the service of forensic investigations to rationalize the use of extended audit procedures at the substantive phase of the audit. In creating the audit evidence needed to rationalize the use of extended procedures for specific accounts under examination the c2 test applied to the N-B profile is a logical choice as it is fundamentally a variance measure for the difference between the Expectation and the Observation. However, there are anomalously sample size issues in that large sample sizes create a proliferation of False Positive Errors—in the audit context this means using extended procedures when in fact they are not likely to be warranted. In this paper we offer a validated sample size range for using the c2 as the N-B screening test. We discriminate tested this suggested sample size range using datasets that conform and that do not conform to the Newcomb-Benford profile. These dataset are detailed in the paper. Finally, we have a Decision Support System that aides in the creation of this c2 screening information. The DSS, programmed an ExcelÔ VBAÒ, is available from the corresponding author as a free download with restriction to its use.



This work is licensed under a Creative Commons Attribution 4.0 License.