Dois ensaios em econometria financeira: modelagem, previsão e stress da estrutura a termo das taxas de juros

• Ano de defesa: 2008
• Autor(a) principal: Rafael Barros de Rezende
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais
• Palavras-chave em Português: Econometria; Finanças Modelos econométricos; Taxa de juros Modelos econométricos
• Link de acesso: https://hdl.handle.net/1843/AMSA-7T6PMU

Introducing a more flexible model of five factors and the first paper of the thesis verifies the in-sample fitting and the out-of-sample forecasting performance of several extensions of the Nelson and Siegel (1987) parametric model which was reinterpreted by Diebold and Li (2006). We used different rules for fixing the parameters /] that govern the models´ exponential components shapes, and predictions were made for different time horizons using different methods. We highlight the Quantile Autoregressive QAR. The results showed that the five factor model presents a superior in-sample fitting, specially in the short and long term maturities of the term structure. Despite this, a greater predictive power is not guaranteed. It is also shown that, depending on the forecasting horizon, different values of the parameters /] can be optimally fixed. We also conclude that the Brazilian term-structure forecasts performed by the QAR estimated at the median are much more accurate than those performed by the autoregressive methods based on mean-regressions. It shows the robustness of the quantile regression models.The purpose of the second paper is to present a framework for the construction of extreme scenarios for the yield curves employing the quantile autoregression model. Through models estimated in the tails of the conditional distributions of the response variable, conditional extreme scenarios for the level and slope factors of the Svensson (1994) model are constructed, enabling the formulation of scenarios for extreme displacements of the yield curves. The proposed framework has proven capable of generating scenarios that simulate the unlikely yield curves movements observed in reality, making the stress tests based on the yield curves less subjective. Thus, financial institutions and regulators can achieve a better interest rate risk management.