On the globalization of stock markets: An application of VECM, SSA technique and mutual information to the G7?

This paper analyzes the process of stock market globalization on the basis of two different approaches: (i) the linear one, based on cointegration tests and vector error correction models (VECM); and (ii) the nonlinear approach, based on Singular Spectrum Analysis (SSA) and mutual information tests. While the cointegration tests are based on regression models and typically capture linearities in the data, mutual information and SSA are well suited for capturing global non-parametric relationships in the data without imposing any structure or restriction on the model. The data used in our empirical analysis were drawn from DataStream and comprise the natural logarithms of relative stock market indexes since 1973 for the G7 countries. The main results point to the conclusion that significant causal effects occur in this context and that mutual information and the global correlation coefficient actually provide more information on this process than VECM, but the direction of causality is difficult to distinguish in the former case. In this field, SSA shows some advantages, since it enabled us to capture the nonlinear causality in both directions. In all cases, however, there is evidence that stock markets are closely related in the long-run over the 36 years analyzed and, in this sense, one may say that they are globalized.