Prediction of chaotic time series by using ANNs, ANFIS and SVMs

摘要

Many biological systems and natural phenomena exhibit chaotic behaviors that are saved into time series data, which can be used to predict their future behavior. Unfortunately, chaotic time series prediction is a challenge and more difficult when the data do not have a similar pattern. In this manner, this work shows the prediction of chaotic time series that have different maximum Lyapunov exponent (MLE) values, which are generated by a chaotic oscillator, and by applying three techniques, namely: artificial neural networks (ANNs), adaptive neuro-fuzzy inference system (ANFIS), and least-square support vector machines (LS-SVM). The predicted chaotic data is compared with respect to the root mean squared error. The three prediction techniques have the same conditions and they are compared with statistical metrics to predict chaotic time series with different MLE values.