Mutual Information Based Input Variable Selection Algorithm and Wavelet Neural Network for Time Series Prediction

Abstract

In this paper we have presented an Integrated Wavelet Neural Network (WNN) model and Mutual Information (MI)-based input selection algorithm for time series prediction. Based on MI the proper input variables, which describe the time series' dynamics properly, will be selected. The WNN Prediction model uses selected variables and predicts the future. This model utilized for time series prediction benchmark in NN3 competition and sunspot data. Comprehensive results show that integrated Mutual information based input variable selection algorithm and wavelet network based prediction model, which uses selected variable from lagged value, outperforms other models in prediction of time series.