Testando a condição descoberta de paridade de juros entre Brasil e Estados Unidos: uma modelagem por meio de GARCH multivariado e volatilidades realizadas

• Ano de defesa: 2017
• Autor(a) principal: Villela, Lucas Moreira
• Orientador(a): Marçal, Emerson Fernandes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Presbiteriana Mackenzie
• Palavras-chave em Português: condição de paridade descoberta de juros; VECM; MGARCH; volatilidade realizada
• Área do conhecimento CNPq: CNPQ::CIENCIAS SOCIAIS APLICADAS::ADMINISTRACAO::ADMINISTRACAO DE EMPRESAS
• Link de acesso: http://dspace.mackenzie.br/handle/10899/23627

This study tests the uncovered interest rate parity between the Brazilian and American markets during the period of June 1986 to August 2016. The validation of the uncovered parity condition implies efficiency between markets. The condition is tested through the VECM methodology proposed in Engle and Granger (1987) utilizing the cointegrating vector testing the uncovered parity on the long term. The Multivariate GARCH model proposed by Bollerslev, Engle and Wooldridge (1988) is used, modeling not only the mean but the variance of the model’s variables, that way controlling the ARCH (autoregressive conditional heteroscedastic) effect in financial series. The variances of the variables are estimated through the Realized Variance estimator, first proposed in Andersen and Bollerslev (1998), in which the authors show it to be a consistent estimate of the integrated variance of a given process. The results validate the uncovered interest parity, showing it to be valid as a long-term equilibrium and that any deviation is corrected in the long term through the exchange rate between Brazil and the United States.