Abstract
The application of recurrence quantification analysis (RQA) and state space divergence reconstruction for the analysis of financial time series in terms of cross-correlation and forecasting is illustrated using high-frequency time series and random heavy-tailed data sets. The results indicate that these techniques, able to deal with non-stationarity in the time series, may contribute to the understanding of the complex dynamics hidden in financial markets. The results demonstrate that financial time series are highly correlated. Finally, an on-line trading strategy is illustrated and the results shown using high-frequency foreign exchange time series.