A Time Series Data Prediction Scheme Using Bilinear Recurrent Neural Network

A time series prediction method based on a BiLinear Recurrent Neural Network (BLRNN) is proposed in this paper. The proposed predictor is based on the BLRNN that has been proven to have robust abilities in modeling and predicting time series. The learning process is further improved by using a multiresolution-based learning algorithm for training the BLRNN so as to make it more robust for the prediction of time series data. The proposed multiresolution-based BLRNN predictor is applied to the long-term prediction of time series data sets. Experiments and results on the Mackey-Glass Series data and Sunspot Series data show that the proposed prediction scheme outperforms both the traditional MultiLayer Perceptron Type Neural Network (MLPNN) and the BLRNN in terms of the normalized mean square error (NMSE).

[1]  N. Stamatis,et al.  Forecasting chaotic cardiovascular time series with an adaptive slope multilayer perceptron neural network , 1999, IEEE Transactions on Biomedical Engineering.

[2]  V. Alarcon-Aquino,et al.  Multiresolution FIR neural-network-based learning algorithm applied to network traffic prediction , 2006, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[3]  G. Nicolis,et al.  Is there a climatic attractor? , 1984, Nature.

[4]  Fionn Murtagh,et al.  A neuro-wavelet strategy for Web traffic forecasting , 1998 .

[5]  Dong-Chul Park,et al.  Complex-bilinear recurrent neural network for equalization of a digital satellite channel , 2002, IEEE Trans. Neural Networks.

[6]  Dong-Chul Park,et al.  Bilinear recurrent neural network , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[7]  A. Babloyantz,et al.  Low-dimensional chaos in an instance of epilepsy. , 1986, Proceedings of the National Academy of Sciences of the United States of America.

[8]  B. LeBaron,et al.  Nonlinear Dynamics, Chaos, and Instability: Statistical Theory and Economic Evidence , 1991 .

[9]  Amir B. Geva,et al.  ScaleNet-multiscale neural-network architecture for time series prediction , 1998, IEEE Trans. Neural Networks.

[10]  Yao Liang,et al.  Multiresolution learning paradigm and signal prediction , 1997, IEEE Trans. Signal Process..

[11]  J. Hale,et al.  Dynamics and Bifurcations , 1991 .

[12]  Mark J. Shensa,et al.  The discrete wavelet transform: wedding the a trous and Mallat algorithms , 1992, IEEE Trans. Signal Process..

[13]  Wen-Rong Wu,et al.  Adaptive AR modeling in white Gaussian noise , 1997, IEEE Trans. Signal Process..

[14]  Fionn Murtagh,et al.  Wavelet-based combined signal filtering and prediction , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[15]  Subbarayan Pasupathy,et al.  Predictive head movement tracking using a Kalman filter , 1997, IEEE Trans. Syst. Man Cybern. Part B.

[16]  Les E. Atlas,et al.  Recurrent neural networks and robust time series prediction , 1994, IEEE Trans. Neural Networks.

[17]  Dong-Chul Park,et al.  Multiscale BiLinear Recurrent Neural Network with an Adaptive Learning Algorithm , 2006, ICNC.

[18]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[19]  Dong-Chul Park,et al.  Multiresolution-Based BiLinear Recurrent Neural Network , 2007, PAKDD.

[20]  Piet M. T. Broersen,et al.  Automatic spectral analysis with time series models , 2002, IEEE Trans. Instrum. Meas..

[21]  Hanoch Lev-Ari,et al.  Optimized estimation of moments for nonstationary signals , 1997, IEEE Trans. Signal Process..