Transform your approach to oprisk modelling with a proven, non-statistical methodology
Operational Risk Modeling in Financial Services provides risk professionals with a forward-looking approach to risk modelling, based on structured management judgement over obsolete statistical methods. Proven over a decade’s use in significant banks and financial services firms in Europe and the US, the Exposure, Occurrence, Impact (XOI) method of operational risk modelling played an instrumental role in reshaping their oprisk modelling approaches; in this book, the expert team that developed this methodology offers practical, in-depth guidance on XOI use and applications for a variety of major risks.
The Basel Committee has dismissed statistical approaches to risk modelling, leaving regulators and practitioners searching for the next generation of oprisk quantification. The XOI method is ideally suited to fulfil this need, as a calculated, coordinated, consistent approach designed to bridge the gap between risk quantification and risk management. This book details the XOI framework and provides essential guidance for practitioners looking to change the oprisk modelling paradigm.
- Survey the range of current practices in operational risk analysis and modelling
- Track recent regulatory trends including capital modelling, stress testing and more
- Understand the XOI oprisk modelling method, and transition away from statistical approaches
- Apply XOI to major operational risks, such as disasters, fraud, conduct, legal and cyber risk
The financial services industry is in dire need of a new standard - a proven, transformational approach to operational risk that eliminates or mitigates the common issues with traditional approaches. Operational Risk Modeling in Financial Services provides practical, real-world guidance toward a more reliable methodology, shifting the conversation toward the future with a new kind of oprisk modelling.
Table of Contents
List of Figures xi
List of Tables xv
Foreword xix
Preface xxi
Part One Lessons Learned in 10 Years of Practice
Chapter 1 Creation of the Method 3
1.1 From Artificial Intelligence to Risk Modelling 3
1.2 Model Losses or Risks? 5
Chapter 2 Introduction to the XOI Method 7
2.1 A Risk Modelling Doctrine 7
2.2 A Knowledge Management Process 8
2.3 The eXposure, Occurrence, Impact (XOI) Approach 9
2.4 The Return of AI: Bayesian Networks for Risk Assessment 10
Chapter 3 Lessons Learned in 10 Years of Practice 13
3.1 Risk and Control Self-Assessment 13
3.2 Loss Data 24
3.3 Quantitative Models 30
3.4 Scenarios Workshops 36
3.5 Correlations 41
3.6 Model Validation 47
Part Two Challenges of Operational Risk Measurement
Chapter 4 Definition and Scope of Operational Risk 57
4.1 On Risk Taxonomies 57
4.2 Definition of Operational Risk 68
Chapter 5 The Importance of Operational Risk 71
5.1 The Importance of Losses 71
5.2 The Importance of Operational Risk Capital 74
5.3 Adequacy of Capital to Losses 76
Chapter 6 The Need for Measurement 77
6.1 Regulatory Requirements 77
6.2 Nonregulatory Requirements 82
Chapter 7 The Challenges of Measurement 93
7.1 Introduction 93
7.2 Measuring Risk or Measuring Risks? 93
7.3 Requirements of a Risk Measurement Method 95
7.4 Risk Measurement Practices 98
Part Three The Practice of Operational Risk Management
Chapter 8 Risk and Control Self-Assessment 105
8.1 Introduction 105
8.2 Risk and Control Identification 107
8.3 Risk and Control Assessment 113
Chapter 9 Losses Modelling 121
9.1 Loss Distribution Approach 122
9.2 Loss Regression 134
Chapter 10 Scenario Analysis 137
10.1 Scope of Scenario Analysis 137
10.2 Scenario Identification 150
10.3 Scenario Assessment 163
Part Four The Exposure, Occurrence, Impact Method
Chapter 11 An Exposure-Based Model 179
11.1 A Tsunami Is Not an Unexpectedly Big Wave 179
11.2 Using Available Knowledge to Inform Risk Analysis 180
11.3 Structured Scenarios Assessment 181
11.4 The XOI Approach: Exposure, Occurrence, and Impact 182
Chapter 12 Introduction to Bayesian Networks 185
12.1 A Bit of History 185
12.2 A Bit of Theory 186
12.3 Influence Diagrams and Decision Theory 187
12.4 Introduction to Inference in Bayesian Networks 187
12.5 Introduction to Learning in Bayesian Networks 189
Chapter 13 Bayesian Networks for Risk Measurement 191
13.1 An Example in Car Fleet Management 191
Chapter 14 The XOI Methodology 203
14.1 Structure Design 203
14.2 Quantification 209
14.3 Simulation 214
Chapter 15 A Scenario in Internal Fraud 219
15.1 Introduction 219
15.2 XOI Modelling 219
Chapter 16 A Scenario in Cyber Risk 227
16.1 Definition 227
16.2 XOI Modelling 234
Chapter 17 A Scenario in Conduct Risk 239
17.1 Definition 239
17.2 Types of Misconduct 241
17.3 XOI Modelling 246
Chapter 18 Aggregation of Scenarios 255
18.1 Introduction 255
18.2 Influence of a Scenario on an Environment Factor 257
18.3 Influence of an Environment Factor on a Scenario 258
18.4 Combining the Influences 261
18.5 Turning the Dependencies into Correlations 262
Chapter 19 Applications 265
19.1 Introduction 265
19.2 Regulatory Applications 267
19.3 Risk Management 278
Chapter 20 A Step towards “Oprisk Metrics” 287
20.1 Introduction 287
20.2 Building Exposure Units Tables 288
20.3 Sources for Driver Quantification 289
20.4 Conclusion 290
Index 291
