Abstract:

Considering the characteristic of chaotic time series that may produce synchronization in the phase space although there is no obvious correlation between the amplitudes of various components, a nonlinear correlation analysis algorithm based on phase space synchronization is presented. Firstly, the nearest neighborhood of one time series is mapping into the phase space of another time series according to the time subscript, and the change of the neighborhood radius is calculated. Then, a criterion is defined to measure the nonlinear dependence between two time series. Finally, a local prediction model is constructed to implement the precise prediction of multivariate time series. Simulation results show the effectiveness of the designed method.