| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.17 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Nesta tese, apresenta-se o método COS: um método numérico para avaliação de opções, baseado na expansão em série de cossenos de Fourier. Foi proposto por Fang e Oosterlee, e pode ser eficazmente utilizado para avaliar opções de estilo Bermudiano subjacentes a processos de Lévy ou ao modelo de volatilidade estocástica de Heston. O ingrediente principal do método é a relação próxima entre a função característica e os coeficientes da expansão em série de cossenos de Fourier da função densidade. Ao considerar o modelo de Heston, o problema de avaliação bidimensional é abordado combinando a ideia anterior com regras de quadratura de ordem superior no domínio da log-variância.
In this thesis, we present the COS method: an option pricing numerical method based on Fourier cosine series expansions. It was proposed by Fang and Oosterlee, and it can be used to effectively price Bermudan options under Lévy processes or the Heston stochastic volatility model. The method’s key ingredient is the close relationship between the characteristic function and the series coefficients of the Fourier cosine expansion of the density function. Under the Heston model, the two-dimensional pricing problem is dealt by combining the prior Fourier cosine series insight with high-order quadrature rules in the log-variance dimension.
In this thesis, we present the COS method: an option pricing numerical method based on Fourier cosine series expansions. It was proposed by Fang and Oosterlee, and it can be used to effectively price Bermudan options under Lévy processes or the Heston stochastic volatility model. The method’s key ingredient is the close relationship between the characteristic function and the series coefficients of the Fourier cosine expansion of the density function. Under the Heston model, the two-dimensional pricing problem is dealt by combining the prior Fourier cosine series insight with high-order quadrature rules in the log-variance dimension.
Descrição
Tese de mestrado em Matemática Financeira, Universidade de Lisboa, Faculdade de Ciências, 2020
Palavras-chave
Exercício antecipado Séries de cossenos de Fourier Transformada rápida de Fourier Processos de Lévy Modelo de Heston Teses de mestrado - 2020
