Modelling and prediction of cyclostationary chaotic time series using periodic autoregressive models

It has been shown that some chaotic time series have cyclostationary characteristics. In this paper, this characteristic is exploited for applications to modeling and prediction of chaotic time series. To this aim, the periodic autoregressive model is used. The application of the proposed model to simulated data from the periodically perturbed logistic map is carried out and the results show that the model works well for modelling and long-term prediction in comparison with other models.