The article "Recurrence quantification analysis of global stock markets", by João A. Bastos and Jorge Caiado, researchers at CEMAPRE- Centro de Matemática Aplicada de Previsão Económica do ISEG, was accepted for publication in the international journal "Physica A: Statistical Mechanics and its Applications". Bastos and Jorge Caiado, researchers at CEMAPRE- Center for Mathematics Applied to Forecasting and Economic Decision of ISEG.
In classical financial theory the dynamics of financial asset prices are typically described by stochastic processes. However, an alternative theory suggests that these prices may be generated by deterministic, non-linear and chaotic processes. Indeed, this type of process reproduces the large fluctuations observed in financial markets that stochastic processes cannot explain. The existence of a deterministic component in price dynamics implies that their evolution is predictable in the short run. In the paper now accepted for publication, the presence of deterministic dependencies in international stock markets is investigated by quantifying recurrences (similar paths taken by a dynamic system at different times) in index prices. The results suggest that the dynamics of emerging market indices may contain a higher deterministic component than those of developed markets. In particular, the dynamics of the US, Japanese and Taiwanese markets appear to have the lowest deterministic component among the 46 markets analyzed. The behavior of some markets during the Asian currency crisis, the dot-com bubble crisis and the recent subprime crisis was also studied. The results suggest that during these periods there was a reduction in the deterministic component of the indices analyzed.
