Publication
Claim Severity modelling in Automotive Insurance: Are Electric Vehicles Riskier?: A case study analysis using Portuguese Data
| Summary: | This dissertation aims to explore the factors influencing the severity of motor insurance claims, with a specific focus on understanding the differences between electric and traditional vehicles. By identifying key variables that impact claim costs, the study seeks to assist insurance companies in making informed decisions to enhance profitability and sustainability. The research employs a combination of statistical analysis and predictive modeling techniques, including Generalized Linear Models (GLM) and Logistic Regression. Data from insurance claims with data related to mandatory third-party liability coverage, segmented into electric and traditional vehicles, are analyzed to identify patterns and variables that significantly affect accident severity. The study reveals distinct characteristics between electric and traditional vehicle claims. Variables such as vehicle age, geographic location, and type of accident contribute differently to the severity of claims in the two segments. For non-electric vehicles, variables such as the vehicle's gross weight, the district, the vehicle's year of construction, driver’s age, years of driving experience and type of vehicle were obtained; for electric vehicles, only the vehicle's year of construction, brand, and the district were found significant. Through the study using logistic regression, we concluded that electric vehicles have a higher probability of causing severe accidents. The results provide actionable insights for insurance companies, enabling them to optimize premium calculations and reduce financial risks associated with claim payouts. By leveraging these findings, insurers can improve their pricing accuracy and competitive positioning in a rapidly evolving market. Understanding the risk profiles of electric and traditional vehicles supports the development of fairer insurance policies. It offers valuable insights for insurers, policymakers, and stakeholders, providing a foundation for more effective risk management and promoting sustainable growth in the motor insurance industry. |
|---|---|
| Main Authors: | Serra, Ana Maria Antunes |
| Subject: | GLM Logistic Regression Non-life Insurance Pricing Severity Predictive Modelling Electric Cars SDG 8 - Decent work and economic growth |
| Year: | 2025 |
| Country: | Portugal |
| Document type: | master thesis |
| Access type: | open access |
| Associated institution: | Universidade Nova de Lisboa |
| Language: | English |
| Origin: | Repositório Institucional da UNL |
| Summary: | This dissertation aims to explore the factors influencing the severity of motor insurance claims, with a specific focus on understanding the differences between electric and traditional vehicles. By identifying key variables that impact claim costs, the study seeks to assist insurance companies in making informed decisions to enhance profitability and sustainability. The research employs a combination of statistical analysis and predictive modeling techniques, including Generalized Linear Models (GLM) and Logistic Regression. Data from insurance claims with data related to mandatory third-party liability coverage, segmented into electric and traditional vehicles, are analyzed to identify patterns and variables that significantly affect accident severity. The study reveals distinct characteristics between electric and traditional vehicle claims. Variables such as vehicle age, geographic location, and type of accident contribute differently to the severity of claims in the two segments. For non-electric vehicles, variables such as the vehicle's gross weight, the district, the vehicle's year of construction, driver’s age, years of driving experience and type of vehicle were obtained; for electric vehicles, only the vehicle's year of construction, brand, and the district were found significant. Through the study using logistic regression, we concluded that electric vehicles have a higher probability of causing severe accidents. The results provide actionable insights for insurance companies, enabling them to optimize premium calculations and reduce financial risks associated with claim payouts. By leveraging these findings, insurers can improve their pricing accuracy and competitive positioning in a rapidly evolving market. Understanding the risk profiles of electric and traditional vehicles supports the development of fairer insurance policies. It offers valuable insights for insurers, policymakers, and stakeholders, providing a foundation for more effective risk management and promoting sustainable growth in the motor insurance industry. |
|---|