Publicação
Genetic algorithms applied to asset & liability management
| Resumo: | Effective asset liability management is at the core of what a life insurance company must do, particularly in what concerns defined benefits pension fund products. The life insurer faces a complex problem whereby multiple and sometimes conflicting objectives must be addressed at the same time, such as achieving higher returns while reducing the portfolio’s exposure to a plethora of risks. To achieve these goals, pension fund managers must then carefully choose asset allocation strategies for their portfolios from an infinite pool of asset combinations and weights. Given the nature of this problem, the use of genetic algorithms seems to be adequate, as this method is particularly well suited to deal with very large and multi-modal solution spaces. The main purpose of this dissertation is to assess how well the genetic algorithm method performs in solving this specific problem, and compare the results with other simpler methods. The results of Genetic Algorithms application were satisfactory and the results of this study suggests that Genetic Algorithms are a useful tool to solve ALM problems. |
|---|---|
| Autores principais: | Almeida, João Filipe Clérigo Ribeiro de |
| Assunto: | Genetic Algorithms Evolutionary Algorithms Pension Funds Life Insurance Asset Liability Management |
| Ano: | 2017 |
| 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: | Effective asset liability management is at the core of what a life insurance company must do, particularly in what concerns defined benefits pension fund products. The life insurer faces a complex problem whereby multiple and sometimes conflicting objectives must be addressed at the same time, such as achieving higher returns while reducing the portfolio’s exposure to a plethora of risks. To achieve these goals, pension fund managers must then carefully choose asset allocation strategies for their portfolios from an infinite pool of asset combinations and weights. Given the nature of this problem, the use of genetic algorithms seems to be adequate, as this method is particularly well suited to deal with very large and multi-modal solution spaces. The main purpose of this dissertation is to assess how well the genetic algorithm method performs in solving this specific problem, and compare the results with other simpler methods. The results of Genetic Algorithms application were satisfactory and the results of this study suggests that Genetic Algorithms are a useful tool to solve ALM problems. |
|---|
Atividades financiadas
Não foram encontrados projetos financiados.