Chaotic Time Series Prediction Using Radial Basis Function Networks
Nguyen Van Truc, Duong Tuan Anh · 2018
Chaotic time series are ubiquitous in several real world areas. But forecasting in chaotic time series is still a challenging task since forecasting power on this kind of time series with some proposed methods is limited. In this work, we propose an efficient method of chaotic time series prediction using Radial Basis Function (RBF) Network. The main idea of our method is the combination of RBF network and chaotic theory, namely phase space reconstruction in order to enhance the prediction quality. In addition, we apply Modified Backpropagation algorithm on training RBF networks which improves the performance of the RBF network remarkably. Experimental results on several real and synthetic datasets of chaotic time series reveal that RBF network with phase space reconstruction outperforms MLP neural networks.