Towards Explainability of Machine Learning Models in Insurance Pricing

Machine learning methods have garnered increasing interest among actuaries in recent years. However, their adoption by practitioners has been limited, partly due to the lack of transparency of these methods, as compared to generalized linear models. In this paper, we discuss the need for model interpretability in property & casualty insurance ratemaking, propose a framework for explaining models, and present a case study to illustrate the framework.

[1]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[2]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[3]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[4]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[5]  S. Mildenhall.  A SYSTEMATIC RELATIONSHIP BETWEEN MINIMUM BIAS AND GENERALIZED LINEAR MODELS , 1999 .

[6]  Cynthia Rudin,et al.  Model Class Reliance: Variable Importance Measures for any Machine Learning Model Class, from the "Rashomon" Perspective , 2018 .

[7]  Kevin Kuo.  DeepTriangle: A Deep Learning Approach to Loss Reserving , 2018, ArXiv.

[8]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[9]  A data driven binning strategy for the construction of insurance tariff classes , 2018 .

[10]  NCCI ’ s 2007 Hazard Group Mapping by , 2009 .

[11]  R. Bailey,et al.  Two Studies in Automobile Insurance Ratemaking , 1960, ASTIN Bulletin.

[12]  Gerda Claeskens,et al.  Unravelling the predictive power of telematics data in car insurance pricing , 2017 .

[13]  Mario V. Wüthrich.  Covariate selection from telematics car driving data , 2017 .

[14]  Bhiksha Raj,et al.  On the Origin of Deep Learning , 2017, ArXiv.

[15]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[16]  D. Hilton.  Conversational processes and causal explanation. , 1990 .

[17]  Andrea Gabrielli,et al.  Neural network embedding of the over-dispersed Poisson reserving model , 2020 .

[18]  Or Biran,et al.  Explanation and Justification in Machine Learning : A Survey Or , 2017 .

[19]  Mario V. Wüthrich.  Machine learning in individual claims reserving , 2016 .

[20]  Guangyuan Gao,et al.  Convolutional Neural Network Classification of Telematics Car Driving Data , 2019 .

[21]  T. VanderWeele,et al.  Covariate Selection , 2022 .

[22]  Donald F. Mango.  An Application of Garne Theory : Property Catastrophe Risk Load by , 1999 .

[23]  Oluwasanmi Koyejo,et al.  Examples are not enough, learn to criticize! Criticism for Interpretability , 2016, NIPS.

[24]  Cynthia Rudin,et al.  All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously , 2019, J. Mach. Learn. Res..

[25]  Andrea Gabrielli,et al.  A NEURAL NETWORK BOOSTED DOUBLE OVERDISPERSED POISSON CLAIMS RESERVING MODEL , 2019, ASTIN Bulletin.

[26]  Hui Zou,et al.  Insurance Premium Prediction via Gradient Tree-Boosted Tweedie Compound Poisson Models , 2015, 1508.06378.

[27]  Tim Miller,et al.  Explainable AI: Beware of Inmates Running the Asylum Or: How I Learnt to Stop Worrying and Love the Social and Behavioural Sciences , 2017, ArXiv.

[28]  Mario V. Wuthrich,et al.  Feature extraction from telematics car driving heatmaps , 2017, European Actuarial Journal.

[29]  Alexander Noll,et al.  Case Study: French Motor Third-Party Liability Claims , 2018 .

[30]  Emil Pitkin,et al.  Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation , 2013, 1309.6392.

[31]  Daniel W. Apley,et al.  Visualizing the effects of predictor variables in black box supervised learning models , 2016, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[32]  Been Kim,et al.  Towards A Rigorous Science of Interpretable Machine Learning , 2017, 1702.08608.

[33]  Pascal Vincent,et al.  Statistical Learning Algorithms Applied to Automobile Insurance Ratemaking , 2003 .

[34]  Tim Miller,et al.  Explanation in Artificial Intelligence: Insights from the Social Sciences , 2017, Artif. Intell..

[35]  Guangyuan Gao,et al.  Claims frequency modeling using telematics car driving data , 2018, Scandinavian Actuarial Journal.

[36]  Guangyuan Gao,et al.  Feature extraction from telematics car driving heatmaps , 2018, European Actuarial Journal.