We develop the SMuRF algorithm that accurately and effectively solves the general problem of convex optimization with a differentiable loss function and multi-type penalty terms. We show in full detail how this algorithm creates sparse models, using varying combinations of Lasso-type penalties, and investigating and documenting all possible model choices.
We adapt tree-based machine learning methods to the problem of insurance pricing, thereby leaving the comfort zone of both traditional ratemaking and machine learning. State-ofthe-art GLMs are compared to regression trees, random forests and gradient boosting machines.
A data set from a Belgian telematics product aimed at young drivers is used to identify how car insurance premiums can be designed based on the telematics data collected by a black box installed in the vehicle.
A fully data driven strategy to incorporate continuous risk factors and geographical information in an insurance tariff.
Telematics technology - the integrated use of telecommunication and informatics - may fundamentally change the car insurance industry by allowing insurers to base their prices on the real driving behavior instead of on traditional policyholder characteristics and historical claims information.