Publicação
Construção de um sistema de bonus-malus na presença de outras variáveis tarifárias
| Resumo: | The main objective of this thesis is the determination of a bonus-malus scale for motor third part liability when a priori risk classification is used, i.e, the bonus-malus system is superimposed on a premium system involving a number of other rating variables. The key idea is that both a priori classification and a posteriori corrections aim to create tariff cells as homogeneous as possible. ln this way the bonus-malus scales are determined in order to avoid the over penalization for bad risks and over benefits for good risks that we see in classics bonus-malus scales. So it can be seen as an extension of well known models of tarification. For this purpose two models are presented: • The first model is a credibility technique that allows us to calculate premiums tables as function of time, past accidents and rating factors. • ln the second modela new bonus-malus system is derived. ln a first stage a regression model is estimate in arder to identify significant risk classification factors, determine tariff class and calculate premiums (a priori model). Given the results of the regression model, a bonus-malus system is estimated (a posteriori model). Finally, in a last chapter, we compare the results and tak:e some conclusions. |
|---|---|
| Autores principais: | Silva, Henda Mondlane Ferreira da |
| Assunto: | Sistema bonus-malus Tarifação a priori Modelo de Poisson Modelo binomial negativo Componente de regressão Credibilidade Bonus-malus system a priori ratemaking Poisson model Negative binomial model Regression component Credibility |
| Ano: | 2006 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | português |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | The main objective of this thesis is the determination of a bonus-malus scale for motor third part liability when a priori risk classification is used, i.e, the bonus-malus system is superimposed on a premium system involving a number of other rating variables. The key idea is that both a priori classification and a posteriori corrections aim to create tariff cells as homogeneous as possible. ln this way the bonus-malus scales are determined in order to avoid the over penalization for bad risks and over benefits for good risks that we see in classics bonus-malus scales. So it can be seen as an extension of well known models of tarification. For this purpose two models are presented: • The first model is a credibility technique that allows us to calculate premiums tables as function of time, past accidents and rating factors. • ln the second modela new bonus-malus system is derived. ln a first stage a regression model is estimate in arder to identify significant risk classification factors, determine tariff class and calculate premiums (a priori model). Given the results of the regression model, a bonus-malus system is estimated (a posteriori model). Finally, in a last chapter, we compare the results and tak:e some conclusions. |
|---|