A STUDY ON FORECASTING FOR TIME SERIES BASED ON WAVELET ANALYSIS
基于小波分解与重构的时间序列预测法

The general forecasting approaches for time series are incapable of forecasting the time series of nonlinear non stationary process. In this paper, an approach based on wavelet decomposition and reconstruction(WDR) is put forward. In comparison with the general forecasting approaches, the forecast accuracy of WDR is improved effectively.