Research report: Machine Learning Methods to Perform Pricing Optimization

As the level of competition increases, pricing optimization is gaining a central role in most mature insurance markets, forcing insurers to optimise their rating and consider customer behaviour; the modeling scene for the latter is one currently dominated by frameworks based on Generalised Linear Models (GLMs). In this paper, the authors explore the applicability of novel machine learning techniques such as tree boosted models to optimise the proposed premium on prospective policyholders. Given their predictive gain over GLMs, the authors carefully analyse both the advantages and disadvatanges induced by their use.

Authors: Giorgio Alfredo Spedicato, Christophe Dutang, and Leonardo Petrini
Publisher: Variance journal
Language: English
Year: 2017
Pages: 21 pages

1 Abstract
2 Introduction
3 Business context overview
4 Predictive modeling for binary classification
4.1 Modeling steps
4.2 Data pre-processing
4.3 Model Training, Tuning and Performance Assessment
4.4 Common predictive models for binary classification
4.5 GLMs and their Elastic Net extension
4.6 Random Forest
4.7 The boosting approach: GBM and XGBoost
4.8 Deep Learning
5 Numerical evidence
5.1 Brief description of the dataset
5.2 Model Comparison and Select
5.3 Application to Pricing Optimization
6 Conclusion
7 Acknowledgements
8 Appendix


0 0 votes
Article Rating
Notify of
Inline Feedbacks
View all comments