In this work we study multiple related (multivariate) time series from worldwide markets. We search for signs of coherence and/or synchronization using the main index as representative of the whole market. In order to better understand the relations between the time series we use two different techniques, entropy and variance-covariance matrices. We apply each procedure in a time dependent way to better understand the underlying dynamics of the system. We found that both methods show that world markets, regardless of their maturity status (mature or emergent), are behaving more and more alike over the last years. The simultaneous use of correlation and entropy to study multivariate time series is a promising approach in the sense that they capture different aspects of the collective system dynamics. Â© 2011 Springer Science+Business Media B.V.
Tipo (Avaliação Docente):