analytic tests.
Fraud and Fraud Detection provides insights that enhance the reader’s data analytic skills. Readers will learn to:
• Understand the different areas of fraud and their specific detection methods.
• Evaluate point-of-sales system data files for zapper fraud.
• Understand data requirements, file format types, and apply data verification procedures.
• Understand and apply statistical sampling techniques.
• Apply correlation and trend analysis to data and evaluate the results.
• Identify anomalies and risk areas using computerized techniques.
• Distinguish between anomalies and fraud.
• Develop a step-by-step plan for detecting fraud through data analytics.
• Utilize IDEA software to automate detection and identification procedures.
The delineation of detection techniques for each type of fraud makes this book a must-have for students and new fraud-prevention professionals, and the step-by-step guidance to automation and complex analytics will prove useful for experienced examiners. With data sets growing exponentially, increasing both the speed and sensitivity of detection helps fraud professionals stay ahead of the game. Fraud and Fraud Detection is a guide to more efficient, more effective fraud identification.
This book is about identifying fraud with the aid of data analytic techniques. It includes some data analytical tests that you probably have not considered. It may also expose you to some useful features of IDEA and how to apply procedures to make you a more effective IDEA user. This is a list of the chapters in Fraud and Fraud Detection:
• Chapter 1, “Introduction.” This chapter provides a simple definition of fraud to distinguish it from abuse. Different types of frauds are outlined. The chapter includes a discussion of why a certain amount of fraud risk is acceptable to organizations and how risk assessments enable us to evaluate and focus on areas with a higher potential risk of fraud.
• Chapter 2, “Fraud Detection.” Occupational fraud is hard to detect as employees know their systems inside out. There are both fraudulent inclusions and fraudulent exclusions to be evaluated. This chapter discusses recognizing the red flags of fraud and different types of anomalies. Accounting and analytic anomalies are distinguished, as well as whether procedures are considered to be data mining or data analytics.
• Chapter 3, “The Data Analysis Cycle.” The data analysis cycle steps include evaluation, technology, and auditing the results from the analysis. Before you can do any analysis, you must have good data. This chapter defines the steps in obtaining the data, such as necessary files, fields, file formats, the download, and techniques to verify data integrity. The next steps of preparing for data analysis include examples of data-familiarization techniques along with arranging and organizing the data.
• Chapter 4, “Statistics and Sampling.” Knowledge of statistics can help to better understand and evaluate anomalies, of which some are deviations from the normal. Results of sampling methods can also be more effectively analyzed and interpreted. This chapter explains basic statistics that can easily be understood by auditors, accountants, and investigators. The chapter includes both descriptive and inferential statistics, the measure of center points, and variability from the center. Standard deviation and its usefulness in comparing data is discussed. Both statistical and nonstatistical sampling methods that include attribute, monetary unit, classical variable, discovery, and random sampling are demonstrated along with explanations of why sampling is not enough.
• Chapter 5, “Data Analytical Tests.” Data analytical tests allow for the detection of anomalies over entire data sets. The tests can evaluate 100 percent of the transactions or records to reduce potentially millions of records to a reasonable amount of high-risk transactions for review. General data analytical tests can be applied in most situations and are ideal as a starting point for an audit when there is no specific fraud or target identified. Benford’s Law for detecting abnormal duplications in data is explained for the primary, advanced, and associated tests. Examples of these are demonstrated using IDEA’s Benford’s Law built-in tests to provide understanding, application, and evaluations of the results. Z-score, relative size factor, number duplication, same-same-same, same-same-different, and even amounts tests are shown with step-by-step instructions of how they can be manually applied using IDEA.
• Chapter 6, “Advanced Data Analytical Tests.” Correlation, trend analysis, and time-series analysis are explained in this chapter and demonstrated using IDEA’s built-in features. Relationship tests known as GEL-1 and GEL-2 are shown with step-by-step instructions of how they can be manually applied using IDEA. The reader will be able to analyze and evaluate the results from these advanced tests.
• Chapter 7, “Skimming and Cash Larceny.” In this chapter, we will look at the differences and similarities of skimming and cash larceny. This type of fraud has fewer data analytical tests that can be performed as the fraud is not recorded in the business system’s databases or is well concealed. Less attention is paid to this area, as the losses are generally not as significant as in other types of fraud. Methods of skimming and larceny are discussed to provide an understanding of these types of frauds. A short case study using two years of sample sales data highlights possible data analytical tests that use charts and pivot tables to provide views for different analytical perspectives.
• Chapter 8, “Billing Schemes.” Billing schemes occur through accounts payable and are centered on trade and expenses payable. The objective for reviewing accounts payable does not stop at looking for fraud by corrupt employees, but it is also useful for detecting errors and inefficiencies in the system. Most of the case study involves real-life payment data of over 2 million records. Data analytical tests from Chapter 5 are applied in practice and demonstrated using the IDEA software. Other specific tests include isolating payments without purchase orders, comparing invoice dates to payment dates, searching the database for post office box numbers, matching employees’ addresses to the suppliers’ addresses, duplicate addresses in the supplier master file, payments made to suppliers not in the master file, and the usage of gap detection on check numbers.
• Chapter 9, “Check Tampering Schemes.” This chapter includes discussions of electronic payment-systems fraud along with traditional check-tampering schemes to better understand both new and old schemes. Methods of obtaining checks and authorized signatures, along with concealment methods, are outlined. Data analytical tests and a short case study that uses sample data are used as examples.
• Chapter 10, “Payroll Fraud.” Some form of payroll fraud happens in almost every organization; whether they are the simple claiming of excess overtime and hours worked or taking complex steps to set up nonexistent employees on the payroll. This chapter outlines data analytical tests to identify anomalies that can be evaluated to determine payroll fraud. The case study uses real payroll payment data to illustrate step-by-step payroll analytical tests. Tests for the payroll register, payroll master, and commission files are discussed.
• Chapter 11, “Expense Reimbursement Schemes.” Travel and entertainment expenses are open to abuse by employees, as improperly reimbursed amounts puts extra money in their pockets. This chapter uses actual travel expense data to highlight analytical tests to identify anomalies that require additional analysis and review. The case study includes methodologies to make the inconsistent
