Scenario Analysis in Risk Management (eBook)

Dr. Bertrand Hassani is Global Head of Research and Innovation - Risk Methodology, Grupo Santander, and Associate Researcher Université Paris 1 Panthéon Sorbonne

Preface 6
Biography 8
Contents 10
1 Introduction 13
1.1 Is this War? 13
1.2 Scenario Planning: Why, What, Where, How, When… 14
1.3 Objectives and Typology 16
1.4 Scenario Pre-requirements 18
1.5 Scenarios, a Living Organism 19
1.6 Risk Culture 20
References 22
2 Environment 23
2.1 The Risk Framework 23
2.2 The Risk Taxonomy: A Base for Story Lines 24
2.3 Risk Interactions and Contagion 26
2.4 The Regulatory Framework 29
References 35
3 The Information Set: Feeding the Scenarios 37
3.1 Characterising Numeric Data 39
3.1.1 Moments 40
3.1.2 Quantiles 41
3.1.3 Dependencies 41
3.2 Data Sciences 42
3.2.1 Data Mining 42
3.2.2 Machine Learning and Artificial Intelligence 44
3.2.3 Common Methodologies 46
References 48
4 The Consensus Approach 50
4.1 The Process 51
4.2 In Practice 54
4.2.1 Pre-workshop 55
4.2.2 The Workshops 56
4.3 For the Manager 57
4.3.1 Sponsorship 58
4.3.2 Buy-In 59
4.3.3 Validation 59
4.3.4 Sign-Offs 59
4.4 Alternatives and Comparison 60
References 61
5 Tilting Strategy: Using Probability Distribution Properties 62
5.1 Theoretical Basis 63
5.1.1 Distributions 63
5.1.2 Risk Measures 67
5.1.3 Fitting 69
5.1.4 Goodness-of-Fit Tests 72
5.2 Application 73
5.3 For the Manager: Pros and Cons 76
5.3.1 Implementation 76
5.3.2 Distribution Selection 77
5.3.3 Risk Measures 77
References 78
6 Leveraging Extreme Value Theory 80
6.1 Introduction 80
6.2 The Extreme Value Framework 82
6.2.1 Fisher–Tippett Theorem 83
6.2.2 The GEV 83
6.2.3 Building the Data Set 85
6.2.4 How to Apply It? 86
6.3 Summary of Results Obtained 88
6.4 Conclusion 90
References 90
7 Fault Trees and Variations 92
7.1 Methodology 93
7.2 In Practice 94
7.2.1 Symbols 94
7.2.2 Construction Steps 97
7.2.3 Analysis 99
7.2.4 For the Manager 100
7.2.5 Calculations: An Example 100
7.3 Alternatives 101
7.3.1 Failure Mode and Effects Analysis 102
7.3.2 Root Cause Analysis 102
7.3.3 Why-Because Strategy 103
7.3.4 Ishikawa's Fishbone Diagrams 104
7.3.5 Fuzzy Logic 105
References 106
8 Bayesian Networks 108
8.1 Introduction 108
8.2 Theory 111
8.2.1 A Practical Focus on the Gaussian Case 114
8.2.2 Moving Towards an Integrated System: Learning 115
8.3 For the Managers 117
References 119
9 Artificial Neural Network to Serve Scenario Analysis Purposes 121
9.1 Origins 122
9.2 In Theory 123
9.3 Learning Algorithms 124
9.4 Application 126
9.5 For the Manager: Pros and Cons 129
References 130
10 Forward-Looking Underlying Information: Working with Time Series 132
10.1 Introduction 132
10.2 Methodology 133
10.2.1 Theoretical Aspects 134
10.2.1.1 Stationary Process 134
10.2.1.2 Autocorrelation 135
10.2.1.3 White Noise 135
10.2.1.4 Estimation 136
10.2.1.5 Seasonality 137
10.2.1.6 Trends 138
10.2.2 The Models 140
10.3 Application 144
References 148
11 Dependencies and Relationships Between Variables 150
11.1 Dependencies, Correlations and Copulas 151
11.1.1 Correlations Measures 151
11.1.2 Regression 153
11.1.3 Copula 160
11.2 For the Manager 164
References 166
Index 168