A Decomposition Prediction Model for Rainfall Time Series and Its Application
Yang Shu-min · China Rural Water and Hydropower · 2007
In order to predict the stochastic time series, the time series is divided into trend component,periodic component and steady random component based on the analysis of its characteristic,respectively.A decomposition prediction model for rainfall is established in this paper.The trend component and period component of the time series are validated and extracted, respectively,and autoregressive model is applied for new artificial serial generation to simulate and predict the steady random component.Applying the model for the rainfall time series prediction,good results are derived,which demonstrates the validity and applicability of the model.