Publicação
Application of Survival Analysis In The Insurance Industry: Predicting Motor Policy Tenure and Unveiling Factors Impacting Retention
| Resumo: | This thesis explores the application of survival analysis in predicting motor policy tenure and identifying factors impacting retention within the insurance industry. Effective customer retention is crucial for insurers to maintain financial stability and growth in a competitive market. Survival analysis models are developed to estimate the probability of a policy to be active over time, allowing for proactive identification of at-risk policies and targeted retention strategies. Key factors influencing policy duration are analyzed using machine learning techniques such as Penalized Cox Proportional Hazards, Random Survival Forest, and Gradient Boosting. Among these methods, Penalized Cox Proportional Hazards is favored for its superior discriminatory power and interpretability. By enhancing understanding of policyholder behavior and improving retention strategies, this research contributes to optimizing business practices and sustaining growth for insurers. The insights gained can inform strategic decisions in customer relationship management, pricing strategies, and service offerings, thereby maximizing customer lifetime value and profitability. |
|---|---|
| Autores principais: | Chaabini, Skander |
| Assunto: | Survival Analysis Motor Insurance Policy Tenure Prediction Penalized Cox Proportional Hazards Customer Retention SDG 8 - Decent work and economic growth SDG 9 - Industry, innovation and infrastructure |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | This thesis explores the application of survival analysis in predicting motor policy tenure and identifying factors impacting retention within the insurance industry. Effective customer retention is crucial for insurers to maintain financial stability and growth in a competitive market. Survival analysis models are developed to estimate the probability of a policy to be active over time, allowing for proactive identification of at-risk policies and targeted retention strategies. Key factors influencing policy duration are analyzed using machine learning techniques such as Penalized Cox Proportional Hazards, Random Survival Forest, and Gradient Boosting. Among these methods, Penalized Cox Proportional Hazards is favored for its superior discriminatory power and interpretability. By enhancing understanding of policyholder behavior and improving retention strategies, this research contributes to optimizing business practices and sustaining growth for insurers. The insights gained can inform strategic decisions in customer relationship management, pricing strategies, and service offerings, thereby maximizing customer lifetime value and profitability. |
|---|