Boosting Poisson regression models with telematics car driving data

With the emergence of telematics car driving data, insurance companies start to boost classical actuarial regression models for claim frequency prediction. In this paper, we propose two data-driven neural network approaches that process telematics car driving data to construct driving behavior risk factors. Neural networks simultaneously accommodate feature engineering and regression modeling. We conclude in our numerical analysis that both classical actuarial risk factors and telematics car driving data is necessary to receive the best predictive models. This indicates that these two sources of information interact and complement each other.