Menu
Home
Log in / Register
 
Home arrow Business & Finance arrow Fraud and fraud detection
< Prev   CONTENTS   Next >

Usage Considerations of Benford's Law

- Users must consider whether a particular data set should be expected to fall into a Benford's Law distribution.

- What test should be run and how should the results of those tests be interpreted?

- When is Benford's Law ineffective?

For effective usage, amounts of less than $10.00 should be removed from the data. IDEA automatically does this task before performing the Benford's Law calculations. The auditor must bear in mind that there are costs associated with false positives (identifying a fraud condition when none is present) as well as with false negatives (failing to identify a fraud condition when one exists). The auditor must consider the level of significance to select for further investigation. A balance must be found so one does not investigate too many or too few transactions.

Are there categories of fraud that cannot be detected? As Benford's Law detects excessive duplications or made-up numbers, it is not appropriate for detecting deletions. If transactions were randomly deleted from sales, say from a point-of-sales system, the randomness would ensure that there would be no variation from the expected distribution of the original data set prior to any deletions.

Conclusion

- Benford's analysis, when used correctly, is a powerful tool for identifying suspect accounts or amounts for further analysis.

- Benford's analysis is a tool to complement additional tests/tools.

- Users have to gain expertise in interpreting results.

Benford's Law is a wonderful tool for initial risk assessment of the contents of a data set. It provides the auditor or investigator with a good starting point. The user must understand the business and the industry to effectively use this tool. Knowledge of the business can quickly eliminate false positives, such as in the example of the automotive manufacturer where sales of cars to dealerships caused spikes for the first two digits in the 18-to-32-digits range, as displayed in Figure 5.3.

 
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Philosophy
Political science
Psychology
Religion
Sociology
Travel