Risco cambial e estrutura a termo da taxa de juros: um estudo com vetores autorregressivos para o Brasil 2002-2014

• Ano de defesa: 2015
• Autor(a) principal: Alexandre Moreira Galvao
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Taxa de juros; Forward premium; Risco país; Estrutura a termo; Forward premium puzzle; Risco cambial; Cambio; Juros Brasil; Taxa de juros Brasil
• Link de acesso: http://hdl.handle.net/1843/BUBD-9W2QJL

The main objective of the present study was to evaluate the currency risk in the Brazilian term structure of interest rate structure by VAR models, between 2002 and 2014, aiming to identify observable, macroeconomic and non-observable variables, in order to substantiate such risk premium and its variations on the interest curve. The currency risk in the Brazilian scenario is obtained from the Forward Premium, which was calculated by the differences on the curves of one-day future DI and the FRA de Cupom Cambial (Forward Rate Agreement of Exchange Coupon). From this difference, we obtain the inflationary expectations among currencies resulting in the exchange premium per maturity given as excess return to be modeled. Two main proposals are presented. In the first model, macroeconomic variables, such as inflation and product, are utilized in addition to uncertainty factors such as country risks and latent factors. Under a restricted arbitrage proposal, a VAR Gaussian state-space model was presented to obtain the non-observable variables in addition to the application of macroeconomic factors. Thus, a structure for understanding the currency risk is provided under different maturities, as applied by Ang and Piazzesi (2003) in the modeling of the term structure of interest rate. It is further understandable that the currency risk is an excess premium in relation to the expected devaluation, and an excess return in the curve of interests. In the second model, the currency risk assessed is based on the measurement of the occurred Forward Premium Puzzle (FPP) as well as in latent factors. It also entails short-term data, which affects the long-term perspective. In conclusion, the structures of short- and long-term can be merged into methods in order to broaden the possibilities of utilizing these estimates. Furthermore, the importance of recent data in the construction of the foreign exchange risk is herein observed, and the point that these models are rooted in the rational expectations theory of the term structure of interest rate is duly noted. Lastly, it has become evident that techniques targeted at the study and modeling of the term structure of interest rate are applicable in the formation of foreign exchange premia, and that these show a forward pattern for related maturity