Relação entre as variáveis macroeconômicas e o retorno acionário nas empresas do setor financeiro dos mercados acionários latino-americano

Abstract

The aim of this study is to verify the relationship between the macroeconomic indicators (Percentage Change in Gross Domestic Product (%ΔPIB), Exchange Rate (TC) and Real Interest Rate (%JR)) with the Return on Equity of the market in five Latin American countries. The data were analyzed from 2010 to 2016 by static panel analysis using three approaches (OLS Pooled, Fixed Effects and Random Effects) and by the Dynamic Method (Generalized Method of Moments), in two approaches, being: Arellano-Bond (GMM-AB) and System (GMM-System). For validation of the proposed model, the assumptions of the multiple regression (unit root test, normality, multicollinearity, heteroscedasticity and autocorrelation) of the model variables. Considering the static panel, the Chow and Hausman tests point out that the most appropriate approach for the sample is Fixed Effects, with time effect. However, for all three approaches, we do not reject H0 for autocorrelation hypothesis, indicating that the dynamic effects estimates may be valid. Therefore, the following tests were performed: first and second order autocorrelation and overidentification. The results show that the share return has positive relation with the exchange rate and negative relation with the GDP. It is worth noting that, using the GMM-AB and GMM-System approach, there is a negative relation between the stock return and the real interest rate, and this relationship only occurred in the GMM-System. In this sense, the results corroborate partially with previous studies.