| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.57 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
The scope of this work is to calculate an Automobile Tariff for Third Party Liability Coverage and Shock, Collision and Rollover Coverage assuming the basic distributions, Poisson for Claim Frequency and Gamma for Average Cost using Generalized Linear Regression Models.
After a brief presentation of the methodology we develop an exploratory analysis of the portfolio both from a frequency and a severity point of view. Using the main conclusions of this exploratory analysis we estimate our models. We use a log link for both frequency and severity, i.e. we define a multiplicative tariff. When modelling the frequency we use a Poisson regression model with over dispersion and we introduce the bonus-malus in force in the company as an offset since the definitions of a new bonus malus system was outside our setup.
Once the models are estimated we define a tariff structure and, finally, in the last chapter we analyze the results and take some conclusions.
Descrição
Mestrado em Ciências Actuariais
Palavras-chave
Tarifa Automóvel Modelos Lineares Generalizados Seguradoras Tarifação Distribuição Poisson Distribuição Gamma Cobertura Responsabilidade Civil Cobertura Danos Próprios Automobile Tariff Generalized Linear Models Gamma distribution Insurance companies Poisson distribution Pricing Collision and Rollover Coverage Third Party Liability Coverage
Contexto Educativo
Citação
Oliveira, Diogo Rocha de (2016). "Ratemaking in motor insurance : an application". Dissertação de Mestrado, Universidade de Lisboa. Instituto Superior de Economia e Gestão.
Editora
Instituto Superior de Economia e Gestão
