European stock markets’ cointegration in the presence of structural breaks (1999-2014)

The main objective of this work is to determine if there is a long-term convergence relationship among the four largest markets in continental Europe. The sample includes the markets of Paris, Frankfurt, Milan and Madrid during a period of important changes in the economic environment and, in particular, episodes of intense turbulence. The methodological approach used consists in the construction of a model capable of representing the levels of the stock markets of the four markets and of characterizing their historical behavior with the use of econometric techniques. The study begins with the confirmation that the behavior thorough time of the stock market indices studied is not stationary in the presence of structural breaks. After confirming the evidence of unitary roots, a cointegration analysis of the natural logarithms of the indices is performed, confirming the existence of, at least, one cointegrating vector. Next, the behavior of the series was modeled with a Vector Errors Correction Model (VECM). The results of the model presented heteroscedasticity problems, so a GARCH family model was used to capture the complexity of the factors that determine that behavior. In effect, once heteroscedasticity was modeled, it is possible to proceed with the interpretation of the coefficients of the model, as well as the nature of the long-term relationship among the series.