Abstract Positive Lyapunov exponents cause the errors in modelling of the chaotic time series to grow exponentially. In this paper, we propose the modified version of the support vector machines (SVM) to deal with this problem. Based on recurrent least squares support vector machines (RLS-SVM), we introduce a weighted term to the cost function to compensate the prediction errors resulting from the positive global Lyapunov exponents. To demonstrate the effectiveness of our algorithm, we use the power spectrum and dynamic invariants involving the Lyapunov exponents and the correlation dimension as criterions, and then apply our method to the Santa Fe competition time series. The simulation results shows that the proposed method can capture the dynamics of the chaotic time series effectively.
Received: 21 April 2004
Revised: 22 September 2004
Accepted manuscript online:
Fund: Project supported by the National Natural Science Foundation of China (Grant No 90207012).
Cite this article:
Sun Jian-Cheng (孙建成), Zhang Tai-Yi (张太镒), Liu Feng (刘枫) Modelling of chaotic systems based on modified weighted recurrent least squares support vector machines 2004 Chinese Physics 13 2045
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.