Chapter 10
Identifying Fraud Using the Largest Subsets and Largest Growth Tests
The Nigrini Cycle looked at the data in a single field only. The data profile and the histogram gave us information on the distribution of the numbers. The periodic chart gave us the monthly totals. The first-order, summation, second-order, and last-two digits tests all analyzed the digit patterns in a single field. The number duplication test also looked at the duplications of specific numbers in a single field. In the parallel scan we also looked at a single field of data but we compared the numerical and graphical statistics over two time periods. In the next four chapters the forensic tests will analyze two or more fields at a time. These tests are drill-down tests that will give us small samples of highly suspicious records. The tests in the next four chapters are called the advanced tests because of the use of more than one field in the analysis.
The first test described is the largest subsets test. The test uses two fields, one with transaction or balance numbers (such as dollars, inventory counts, vote counts, population counts) and another field to indicate the subset (e.g., vendor number, credit card number, or branch number). The word subset is borrowed from mathematics where we know a set to be a collection of distinct objects and a subset to be a set whose members are all contained in another set. In our case our subset is a group of records that have something in common. This could be ...
