Case Study: French Motor Third-Party Liability Claims
Authors:
Preprint
- 18 April 2018
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
We provide a case study that compares a classical generalized linear model for claims frequency modeling to regression tree, boosting machine and neural network approaches. We explore these methods, discuss their calibration and study their predictive power on an explicit motor third-party liability insurance data set. The results of the case study show that a simple generalized linear model does not capture interactions of feature components appropriately, whereas the other methods are able to address these interactions.Keywords
This publication has 22 references indexed in Scilit:
- On Boosting: Theory and ApplicationsSSRN Electronic Journal, 2019
- Insights from Inside Neural NetworksSSRN Electronic Journal, 2018
- Delta Boosting Machine with Application to General InsuranceNorth American Actuarial Journal, 2018
- A data driven binning strategy for the construction of insurance tariff classesScandinavian Actuarial Journal, 2018
- Other Neural Networks for Deep Learning2016
- Deep learningNature, 2015
- A Decision-Theoretic Generalization of On-Line Learning and an Application to BoostingJournal of Computer and System Sciences, 1997
- Cryptographic limitations on learning Boolean formulae and finite automataJournal of the ACM, 1994
- Approximation by superpositions of a sigmoidal functionMathematics of Control, Signals, and Systems, 1989
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989