Study Of UBI Vehicle Insurance Pricing Strategy Based On Driving Behavior Data Analysis

Vehicle traffic accidents are largely caused by the driver's improper driving behavior.However,the impact of driving behavior to the accident is not considered by the traditional vehicles insurance premium pricing model,resulting in unreasonable vehicle insurance premium pricing.With the rise of Internet of Vehicles and big data analysis,the real-time behavioral data of vehicles and drivers can be collected by vehicle-mounted sensors and uploaded to the data center,providing a basis and reference for vehicle premium pricing.The new pricing model of UBI(Usage-Based Insurance)Based on driving behavior analysis is gradually coming into public view.Based on the actual vehicle measurement data of an insurance company,this paper studies the UBI premium pricing strategy based on driving behavior analysis.The research content mainly includes three parts:(1)Using the generalized linear model,the risk probability of vehicles was analyzed and predicted by the logical regression model,the degree of influence of different driving behavior data on vehicle risk was studied,the driving behavior risk assessment model was established,and the corresponding vehicle insurance premium was formulated.(2)Through the compound poisson-gamma distribution of Tweedie distribution family,the accumulated claim amount of vehicles insurance was studied and analyzed,and the vehicle pure premium prediction model based on driving behavior was established,which was used as pricing guidance and reference for vehicle insurance premiums.(3)Using the GAMLSS,through zero adjustment inverse gaussian distribution,the amount of vehicle insurance claims was studied and analyzed through parameters of location,scale and shape,and the vehicle pure premium prediction model was established as the reference for vehicle insurance premium pricing,which has a better fitting effect than the Tweedie distribution family prediction model.