Wavelet-based analysis and forecasting for non-stationary time series
Ma She · Journal of systems engineering · 2000
This paper introduced a method of wavelet based analysis and forecasting for non stationary time series. By wavelet decomposing, time series is decomposed into many series accord ing to scale. Trend term, cycle term and stochastic term are separated from original time series in this way. By building model and forecasting for every series and composing the result, we get the forecasting of original time series. The method is feasible through two examples test.