A novel multi-step adaptive prediction method for chaotic time series

Abstract

Based on the short-term predictability of chaotic time series and the adaptive tracking chaotic trajectory of adaptive algorithm, a novel multi-step adaptive prediction method is proposed in this paper to resolve the problem of adjusting the filter parameters of the local adaptive prediction method during multi-step prediction. Simulation results show that this multi-step adaptive prediction method can be successfully used to make multi-step predictions, and its performance is better than that of the local adaptive prediction algorithm.