Claims Frequency Modeling Using Telematics Car Driving Data

Scandinavian Actuarial Journal, Forthcoming

23 Pages Posted: 23 Jan 2018 Last revised: 23 Sep 2018

See all articles by Guangyuan Gao

Guangyuan Gao

Renmin University of China - School of Statistics

Shengwang Meng

School of Statistics, Renmin University of China

Mario V. Wuthrich

RiskLab, ETH Zurich

Date Written: January 15, 2018

Abstract

We investigate the predictive power of covariates extracted from telematics car driving data using the speed-acceleration heatmaps of Gao and Wüthrich (2017). These telematics covariates include K-means classication, principal components, and bottleneck activations from a bottleneck neural network. In the conducted case study it turns out that the first principal component and the bottleneck activations give a better out-of-sample prediction for claims frequencies than other traditional pricing factors such as driver's age. Based on these numerical examples we recommend the use of these telematics covariates for car insurance pricing.

Keywords: Telematics Data, K-Means Algorithm, Principal Components Analysis, Bottleneck Neural Network, Generalized Additive Model, v-a Heatmap, Pattern Recognition, Kullback-Leibler Divergence, Claims Frequency Modeling, Car Insurance Pricing

JEL Classification: G22

Suggested Citation

Gao, Guangyuan and Meng, Shengwang and Wuthrich, Mario V., Claims Frequency Modeling Using Telematics Car Driving Data (January 15, 2018). Scandinavian Actuarial Journal, Forthcoming. Available at SSRN: https://ssrn.com/abstract=3102371 or http://dx.doi.org/10.2139/ssrn.3102371

Guangyuan Gao

Renmin University of China - School of Statistics ( email )

No.59 Zhongguancun Street, Renmin University
Beijing, 100872
China

Shengwang Meng

School of Statistics, Renmin University of China ( email )

Beijing, 100872
China

Mario V. Wuthrich (Contact Author)

RiskLab, ETH Zurich ( email )

Department of Mathematics
Ramistrasse 101
Zurich, 8092
Switzerland

Register to save articles to
your library

Register

Paper statistics

Downloads
321
rank
91,844
Abstract Views
901
PlumX Metrics