Prediction of City Short Term Traffic Flow based on Wavelet Transformation and GM-ARMA Model

A traffic flow prediction method based on wavelet transformation and GM-ARMA model was proposed. By multi-scale wavelet decomposition, the data of traffic flow were decomposed into a low frequency signal and several high frequency signals. Then reconstruct the low frequency and high frequency separately. The low frequency was predicted with grey model and the high frequency signals were predicted with ARMA model. Hence, the prediction value of original sequence was the composition of the respective prediction. The experimental result shows that this method has a higher accuracy than other ones.