Publicação

Price elasticity in motor insurance: Defining price elasticity with causal inference for a major Portuguese insurer

Ver documento

Detalhes bibliográficos
Resumo:Demand-based pricing allows insurance companies to improve their profitability by setting their prices accordantly with customer’s willingness to pay. To employ such price strategy a reliable estimation of customers’ price elasticity is paramount. This dissertation explores the construction of price elasticity functions by applying a causal inference framework which permits reducing the risks of extrapolation of price changes effects on an observational study. The causal inference framework presented leverages from a matching technique which restricts the inference of alternative price changes effect on churn from comparable customers only. The price elasticity functions were estimated by policy degree of coverage since we believe that price sensitivity between Third-party liability (TPL) and Own damage (OD) policies are very distinct. As result, we were able to draw the price elasticity functions for specific subpopulations of customers based on tenure, premium and the existence of claims in the past year.
Autores principais:Lince, Miguel Ângelo Nunes
Assunto:Casual inference Price elasticity Churn prediction Machine Learning Motor insurance Pricing optimization SDG 8 - Decent work and economic growth SDG 9 - Industry, innovation and infrastructure SDG 11 - Sustainable cities and communities
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
Descrição
Resumo:Demand-based pricing allows insurance companies to improve their profitability by setting their prices accordantly with customer’s willingness to pay. To employ such price strategy a reliable estimation of customers’ price elasticity is paramount. This dissertation explores the construction of price elasticity functions by applying a causal inference framework which permits reducing the risks of extrapolation of price changes effects on an observational study. The causal inference framework presented leverages from a matching technique which restricts the inference of alternative price changes effect on churn from comparable customers only. The price elasticity functions were estimated by policy degree of coverage since we believe that price sensitivity between Third-party liability (TPL) and Own damage (OD) policies are very distinct. As result, we were able to draw the price elasticity functions for specific subpopulations of customers based on tenure, premium and the existence of claims in the past year.