DATA ANALYSIS USES TECHNOLOGY to detect anomalies, patterns, and risk indicators within the data set. It can be used to establish a hypothesis or to quantify detected issues if the hypothesis was found to correctly identify fraud.
The true power of data analytics is that the entire data set of the transactions can be tested. Unlike sampling where only a part of the population is tested, data analytics can test 100 percent of the transactions. Resulting anomalies can then all be reviewed or, if in large quantities, sampled.
While the analysis can provide a list of anomalies, it is not a list of fraudulent transactions. Unlike statistical sampling, there is no mathematical formula that provides the auditor with a listing of frauds.
The auditor needs to apply professional judgment, employ analytical skills, and use intuition. Typically, the auditor reviews the list of anomalies, audits some of the transactions, revises the hypothesis, adjusts the test, and performs additional analytical procedures to refine the list to reduce false-positive transactions. There will be numerous false positives of true data anomalies that are not fraud. This is a product of data analytics.
The circular process may continue several times. When completed, the test identifies transactions with a high risk of fraud. This manageable number of transactions can then be examined using fraud-audit procedures. Once a single fraudulent transaction is detected, the audit plan ...