CHAU CHAN YIP (DEREK) is Principal Technical Consultant at SAS Hong Kong since 2010. He was formerly a Technology Consultant at Hewlett Packard. He specializes in data integration and anti-money laundering. He received his Master of Science degree in Computer Science from the Chinese University of Hong Kong.

MAARTEN VAN DIJCK NEMCSIK, LLM, PHD, has worked with SAS since 2012 as a financial crime and tax compliance domain expert and solution lead. He is part of the SAS Global Fraud & Security Business Intelligence Unit, responsible for internal and external training courses in the financial crime and compliance space.

Prologue

Acknowledgments

1              AN INTRODUCTION TO ANTI-MONEY LAUNDERING           4

1.1          THE EMERGENCE OF AML             4

1.2          AML AS A COMPLIANCE DOMAINDOMAIN            5

1.3          THE OBJECTIVES OF AML               7

1.3.1      Regulatory Reporting     7

1.3.2      Corporate Citizenship versus Profitability               7

1.3.3      About true and false positives and negatives       8

1.4          THE EVOLUTION OF AUTOMATED TRANSACTION MONITORING  11

1.4.1      From Rule-Based to Risk-Based  11

1.4.2      From Static to More Dynamic Transaction Monitoring      14

1.4.3      Latest Trends: Machine Learning and Artificial Intelligence            16

1.4.4      Latest Trends: Blockchain             18

1.4.5      Risk-based granularity and statistical relevance  20

1.4.6      Summary             20

2              TRANSACTION MONITORING IN DIFFERENT BUSINESSES 22

2.1          BANKING            23

2.2          CORRESPONDENT BANKING        24

2.3          BANKING – TRADE FINANCE        26

2.4          BANKING – CREDIT CARD              31

2.5          INSURANCE        32

2.6          SECURITIES         33

2.7          STORED VALUE FACILITIES (SVFS)              34

2.8          CASINOS AND ONLINE GAMBLING            36

2.9          LOTTERY AND JOCKEY CLUB         37

2.10        OTHER BUSINESSES         37

2.11        SUMMARY          38

3              THE ALL-IMPORTANCE OF DATA 39

3.1          ETL         39

3.2          EXTRACT: DATA AVAILABILITY AND SOURCING    39

3.3          TRANSFORM: DATA QUALITY, CONVERSION AND REPAIR 41

3.4          DATA LOAD AND FURTHER PROCESSING 46

3.4.1      Loading of the data         46

3.4.2      Data Lineage      48

3.5          MULTIPLE ETLS  48

3.6          SUMMARY          48

4              TYPICAL SCENARIO ELEMENTS    51

4.1          TRANSACTION TYPES      51

4.2          ACTIONABLE ENTITY       53

4.3          SCENARIO PARAMETERS               57

4.3.1      Use of Maximum instead of Minimum Value threshold   58

4.3.2      Threshold per Customer               59

4.4          PRE-COMPUTING DATA 59

4.5          TIMELINESS OF ALERTS  61

4.6          USE OF RATIOS  62

4.6.1      Ratio as Degree of Change / Similarity     62

4.6.2      Ratio as Proportion         64

4.6.3      Other Common Issues   64

5              SCENARIOS IN DETAIL     65

5.1          LARGE AGGREGATE VALUE          65

5.2          UNEXPECTED TRANSACTION       65

5.3          HIGH VELOCITY / TURNOVER      69

5.4          TURNAROUND / ROUND TRIPPING           70

5.5          STRUCTURING  74

5.6          EARLY TERMINATION / QUICK RECIPROCAL ACTION          76

5.7          WATCH LIST        76

5.8          COMMON SPECIFICATION ACROSS UNRELATED ENTITIES               77

5.9          INVOLVE UNRELATED THIRD PARTY          77

5.10        ONE TO MANY  78

5.11        TRANSACTING RIGHT BELOW REPORTING THRESHOLD    78

6              THE SELECTION OF SCENARIOS  80

6.1          SELECTING SCENARIOS  80

6.2          REGULATORY REQUIREMENT      80

6.3          BUSINESS DRIVERS          80

6.4          DATA QUALITY AND AVAILABILITY OF REFERENCE DATA  81

6.5          MAINTENANCE OF THE SCENARIOS REPOSITORY 82

6.6          HOW SPECIFIC A SCENARIO RULE SHOULD BE?    82

6.7          INTER-COVERING SCENARIO RULES          83

6.8          SUMMARY          84

7              ENTITY RESOLUTION AND WATCHLIST MATCHING             85

7.1          ENTITY RESOLUTION       85

7.2          WATCHLISTS      86

8              CUSTOMER SEGMENTATION      101

8.1          THE NEED FOR SEGMENTING CUSTOMERS           101

8.2          APPROACHES TO SEGMENTATION            102

8.3          OVERVIEW OF SEGMENTATION STEPS    104

8.4          ORGANIZATIONAL PROFILING    105

8.5          COMMON SEGMENTATION DIMENSIONS              105

8.6          CONSIDERATIONS IN DEFINING SEGMENTS          106

8.7          CHECK SOURCE DATA FOR SEGMENTATION         107

8.8          VERIFY WITH STATISTICAL ANALYSIS        108

8.9          ONGOING MONITORING              111

8.10        CHANGE OF SEGMENTATION      112

9              SCENARIO THRESHOLD TUNING 114

9.1          THE NEED FOR TUNING 114

9.1.1      Parameters and Thresholds         114

9.1.2      True versus False, Positive versus Negative          115

9.1.3      Cost       115

9.1.4      116

9.1.5      Adapting to the Environment     116

9.1.6      Relatively Simple Ways to Tune Thresholds          117

9.2          OBJECTIVE OF SCENARIO THRESHOLD TUNING    117

9.2.1      Increasing alert productivity        117

9.2.2      Definition of a productive alert   118

9.3          USE OF THRESHOLDS IN DIFFERENT KINDS OF SCENARIO RULES   119

9.3.1      Regulation-driven rules 119

9.3.2      Statistical Outlier              120

9.3.3      122

9.3.4      Insignificance Threshold 122

9.3.5      Safety-Blanket Rules      122

9.4          COMBINING PARAMETERS          123

9.5          STEPS FOR THRESHOLD TUNING 125

9.6          PREPARATION OF ANALYSIS DATA            127

9.6.1      Scope of Data    127

9.6.2      Data Columns    128

9.6.3      Quick and easy approach              129

9.6.4      Analysis of Dates              129

9.6.5      Stratified Sampling          130

9.7          STATISTICAL ANALYSIS OF EACH TUNABLE SCENARIO THRESHOLD VARIABLE          130

9.7.1      Population Distribution Table by Percentile (Ranking analysis)      133

9.8          DISTRIBUTION DIAGRAM COMPRESSED AS A SINGLE LINE              134

9.8.1      Multiple peaks  135

9.8.2      Zeros     136

9.9          ABOVE-THE-LINE ANALYSIS AND BELOW-THE-LINE ANALYSIS        136

9.9.1      Above-the-line analysis 136

9.9.2      Below-the-line analysis 137

9.9.3      Use of Scatter Plots and Interactions between Parameter Variables          139

9.9.4      Binary Search     147

9.10        WHAT-IF TESTS AND MOCK INVESTIGATION         148

9.10.1    What-if Tests     148

9.10.2    Sample Comparisons of What-if Tests     149

9.10.3    Qualifying Results of what-if tests             149

9.11        SCENARIO REVIEW REPORT          150

9.11.1    Scenario Review approach           153

9.11.2    Scenario Review Results 154

9.12        SUMMARY          157

About the Authors

Index

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.