Cointegration and causality analysis of macroeconomic variables and Dow Jones on Ibovespa

Abstract

This article aims to analyze the degree of causality and cointegration between a set of macroeconomic variables, expressed by the Selic rate, exchange rate and the index of industrial production jointly to the Dow Jones index on the Bovespa. The period selected for investigation to the months from January 1995 to December 2012. Econometric model used is the method of Vector Auto Regression with Error Correction (VEC). The unit root tests indicate that the series are integrated of order I (1). In the analysis of the VEC of the adjustment parameters was statistically significant indicating that the Bovespa reacts in the trajectory of long-term equilibrium to changes in the short term. On Decomposition of Forecast Error Variance of the results show the explanatory power of the Bovespa index on its own variance.