A carregar...
Publicação
Seguro de Saúde Individual Variáveis Diferenciadoras do Risco na cobertura de Internamento
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.39 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
O financiamento dos cuidados de saúde em Portugal é assegurado pelo Serviço Nacional de Saúde (SNS); no entanto, para um acesso mais rápido, com mais opção de escolha e comodidade, os portugueses podem recorrer a um seguro de saúde. Para isso, terão que pagar um prémio que é definido de forma idêntica para pessoas com fatores de risco semelhantes. O objetivo deste trabalho é estudar que variáveis, dentro de um conjunto de características pessoais, do seguro escolhido, socioeconómicas e de saúde, têm impacto no risco, ou seja, na frequência de sinistralidade e no custo associado.
O objetivo desta dissertação consiste na identificação de variáveis diferenciadoras do risco da cobertura de Internamento, para os seguros de saúde individuais. O modelo de risco que identifica as variáveis significativas assenta nos Modelos Lineares Generalizados e será analisado o comportamento do risco das variáveis significativas através de duas técnicas de Machine Learning: Análise de Clusters e Árvores de Decisão. Os dados da carteira foram fornecidos por uma Seguradora a operar em Portugal.
The financing of health care in Portugal is provided by the National Health Service (SNS); however, for faster access, with more choice and convenience, the Portuguese can take out health insurance. For this, they will have to pay a premium that is defined identically for people with similar risk factors. The objective of this work is to study which variables, within a set of personal characteristics, of the chosen insurance, socio-economic and health, have an impact on risk, that is, on the frequency of accidents and the cost associated with this loss. The objective of this dissertation is to identify variables differentiating the risk of hospitalization coverage for individual health insurance. The risk model that identifies the significant variables is based on the Generalized Linear Models and the risk behavior of the significant variables will be analyzed through two machine learning techniques: Cluster Analysis and Decision Trees. The portfolio data were provided by an Insurance Company operating in Portugal.
The financing of health care in Portugal is provided by the National Health Service (SNS); however, for faster access, with more choice and convenience, the Portuguese can take out health insurance. For this, they will have to pay a premium that is defined identically for people with similar risk factors. The objective of this work is to study which variables, within a set of personal characteristics, of the chosen insurance, socio-economic and health, have an impact on risk, that is, on the frequency of accidents and the cost associated with this loss. The objective of this dissertation is to identify variables differentiating the risk of hospitalization coverage for individual health insurance. The risk model that identifies the significant variables is based on the Generalized Linear Models and the risk behavior of the significant variables will be analyzed through two machine learning techniques: Cluster Analysis and Decision Trees. The portfolio data were provided by an Insurance Company operating in Portugal.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Risk Analysis and Management
Palavras-chave
Variáveis Diferenciadoras do Risco do Internamento Seguro de Saúde Modelos Lineares Generalizados Análise de Clusters Árvores de Decisão Risk Differentiating Variables of Hospitalization Health insurance Generalized Linear Models Cluster Analysis Decision Trees