A demanda por moeda no Brasil no período 1974-2008: estimativas sob expectativas adaptativas e racionais

• Autor(a) principal: Modesto, Lidiane de Almeida
• Data de Publicação: 2010
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/7576
• topic: Demand for money; The Cagan Model; Adaptative expectations; Box-Cox; Rational expectations; Demanda por moeda; Modelo de Cagan; Expectativas adaptativas; Box-Cox; Expectativas racionais; CNPQ::CIENCIAS SOCIAIS APLICADAS::DEMOGRAFIA::POLITICA PUBLICA E POPULACAO

The main goal of this thesis is analyze the demand for money in Brazil for the period of 1974-2008, noting that it includes sub-periods of high and low inflation, assuming alternative hypothesis for the formation of expectations. The specification adopted is from Tourinho (1995), which generalizes Cagan (1956) to allow a more flexible functional form and incorporate other variables, besides the expected inflation as explanatory variables. This extensions are important to model the demand for real monetary balances in the period considered here. The semi-log functional form of Cagan is rejected in favor of a flexible functional form Box-Cox, and the coefficients of the real interest rate and the variance of inflation are significant, showing the importance of these variables are inserted into the model. The function estimated for the full period is compared with those estimated for the sub-periods of high and low inflation to verify the stability of the formulation adopted. It follows that we can reject the hypothesis that it is stable. Cagan's model is generalized here in another dimension, considering alternative mechanisms for the formation of expectations, which may be adaptive, as the original model, or rational. The hypothesis that adaptive expectations are rational is also considered. The conclusion is that the imposition of the condition of rationality to the model with adaptive expectations does not produce significant changes in the estimated values.