Uma aplicação de métodos de renormalização ao estudo de séries temporais financeiras

• Ano de defesa: 2009
• Autor(a) principal: Renato Vieira dos Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Hipótese de mercado eficiente; Método Detrended Fluctuation Analysis (DFA); Aproximação auto-similar; Probabilidade; Séries temporais; Sistemas dinâmicos; Exponente de Hurst; Renormalização; Física
• Link de acesso: http://hdl.handle.net/1843/ESCZ-7YUHUH

In this work we perform a study that compares, in the light of the Efficient Market Hypothesis, a traditional technique used in the study of time series with a technique motivated by renormalization methods of theoretical physics. We use the technique called the Self-Similar Aproximation to predict of some financial indices for the period that corresponds to the recent global financial crisis of 2008. A review of some concepts of systems was necessary for a better understandingof the proposed method was achieved. We also found the probability distributions that describe the log-price returns. Using the Detrended Fluctuation Analysis (DFA), we calculate the value of the Hurst exponentof some series of the Bovespa index in order to determine possible long-term trends and, in some way, justify the use of the renormalization method proposed.