Big Data for Insurance Companies. Edition No. 1

Table of Contents

Foreword xi
Jean-Charles POMEROL

Introduction  xiii
Marine CORLOSQUET-HABART and Jacques JANSSEN

Chapter 1. Introduction to Big Data and Its Applications in Insurance 1
Romain BILLOT, Cécile BOTHOREL and Philippe LENCA

1.1. The explosion of data: a typical day in the 2010s 1

1.2. How is big data defined?  4

1.3. Characterizing big data with the five Vs  5

1.3.1. Variety  6

1.3.2. Volume  7

1.3.3. Velocity   9

1.3.4. Towards the five Vs: veracity and value 9

1.3.5. Other possible Vs 11

1.4. Architecture 11

1.4.1. An increasingly complex technical ecosystem 12

1.4.2. Migration towards a data-oriented strategy 17

1.4.3. Is migration towards a big data architecture necessary? 18

1.5. Challenges and opportunities for the world of insurance  20

1.6. Conclusion  22

1.7. Bibliography   23

Chapter 2. From Conventional Data Analysis Methods to Big Data Analytics 27
Gilbert SAPORTA

2.1. From data analysis to data mining: exploring and predicting  27

2.2. Obsolete approaches 28

2.3. Understanding or predicting? 30

2.4. Validation of predictive models 30

2.4.1. Elements of learning theory 31

2.4.2. Cross-validation  34

2.5. Combination of models 34

2.6. The high dimension case 36

2.6.1. Regularized regressions 36

2.6.2. Sparse methods  38

2.7. The end of science? 39

2.8. Bibliography   40

Chapter 3. Statistical Learning Methods  43
Franck VERMET

3.1. Introduction  43

3.1.1. Supervised learning 44

3.1.2. Unsupervised learning  46

3.2. Decision trees  46

3.3. Neural networks  49

3.3.1. From real to formal neuron 50

3.3.2. Simple Perceptron as linear separator  52

3.3.3. Multilayer Perceptron as a function approximation tool 54

3.3.4. The gradient backpropagation algorithm 56

3.4. Support vector machines (SVM)  62

3.4.1. Linear separator  62

3.4.2. Nonlinear separator 66

3.5. Model aggregation methods 66

3.5.1. Bagging  67

3.5.2. Random forests  69

3.5.3. Boosting   70

3.5.4. Stacking   74

3.6. Kohonen unsupervised classification algorithm  74

3.6.1. Notations and definition of the model  76

3.6.2. Kohonen algorithm 77

3.6.3. Applications  79

3.7. Bibliography   79

Chapter 4. Current Vision and Market Prospective 83
Florence PICARD

4.1. The insurance market: structured, regulated and long-term perspective 83

4.1.1. A highly regulated and controlled profession 84

4.1.2. A wide range of long-term activities  85

4.1.3. A market related to economic activity  87

4.1.4. Products that are contracts: a business based on the law  87

4.1.5. An economic model based on data and actuarial expertise  88

4.2. Big data context: new uses, new behaviors and new economic models 89

4.2.1. Impact of big data on insurance companies  90

4.2.2. Big data and digital: a profound societal change 91

4.2.3. Client confidence in algorithms and technology 93

4.2.4. Some sort of negligence as regards the possible consequences of digital traces   94

4.2.5. New economic models  95

4.3. Opportunities: new methods, new offers, new insurable risks, new management tools  95

4.3.1. New data processing methods  96

4.3.2. Personalized marketing and refined prices 98

4.3.3. New offers based on new criteria  100

4.3.4. New risks to be insured 101

4.3.5. New methods to better serve and manage clients 102

4.4. Risks weakening of the business: competition from new actors, “uberization”, contraction of market volume 103

4.4.1. The risk of demutualization 103

4.4.2. The risk of “uberization” 104

4.4.3. The risk of an omniscient “Google” in the dominant position due to data 105

4.4.4. The risk of competition with new companies created for a digital world 105

4.4.5. The risk of reduction in the scope of property insurance  106

4.4.6. The risk of non-access to data or prohibition of use  107

4.4.7. The risk of cyber attacks and the risk of non-compliance  108

4.4.8. Risks of internal rigidities and training efforts to implement  109

4.5. Ethical and trust issues 109

4.5.1. Ethical charter and labeling: proof of loyalty 110

4.5.2. Price, ethics and trust 112

4.6. Mobilization of insurers in view of big data  113

4.6.1. A first-phase “new converts”  113

4.6.2. A phase of appropriation and experimentation in different fields  115

4.6.3. Changes in organization and management and major training efforts to be carried out  118

4.6.4. A new form of insurance: “connected” insurance 118

4.6.5. Insurtech and collaborative economy press for innovation  121

4.7. Strategy avenues for the future 122

4.7.1. Paradoxes and anticipation difficulties  122

4.7.2. Several possible choices 123

4.7.3. Unavoidable developments 127

4.8. Bibliography   128

Chapter 5. Using Big Data in Insurance  131
Emmanuel BERTHELÉ

5.1. Insurance, an industry particularly suited to the development of big data 131

5.1.1. An industry that has developed through the use of data  131

5.1.2. Link between data and insurable assets  136

5.1.3. Multiplication of data sources of potential interest  138

5.2. Examples of application in different insurance activities  141

5.2.1. Use for pricing purposes and product offer orientation   142

5.2.2. Automobile insurance and telematics  143

5.2.3. Index-based insurance of weather-sensitive events 145

5.2.4. Orientation of savings in life insurance in a context of low interest rates 146

5.2.5. Fight against fraud 148

5.2.6. Asset management 150

5.2.7. Reinsurance  150

5.3. New professions and evolution of induced organizations for insurance companies  151

5.3.1. New professions related to data management, processing and valuation 151

5.3.2. Development of partnerships between insurers and third-party companies 153

5.4. Development constraints 153

5.4.1. Constraints specific to the insurance industry 153

5.4.2. Constraints non-specific to the insurance industry  155

5.4.3. Constraints, according to the purposes, with regard to the types of algorithms used  158

5.4.4. Scarcity of profiles and main differences with actuaries  159

5.5. Bibliography   161

List of Authors  163

Index 165

Authors