Dynamic fluctuations of cross-correlations in multi-time scale

Abstract With the empirical mode decomposition, we investigate the dynamic fluctuations of cross-correlations in multi-time scale. In this approach, the time series of price returns in financial markets is decomposed into a small number of intrinsic mode functions, and the corresponding time series of phases can be obtained with the Hilbert transform. Through the empirical mode decomposition, we uncover that the cross-correlations between stocks change significantly for different time scales and exhibit singular behaviors during a financial crisis. With the phase correlation analysis, we observe that the cross-correlations between market indices vary also for different time scales. Dividing the total time interval into two sub-intervals by the year 2007, we detect that the market indices are much more correlated after 2007.