A combined forecasting method integrating contextual knowledge

Abstract

According to Qian's meta-synthesis theory and TEI@I methodology,this paper proposes a combined forecasting method based on integrated contextual knowledge(CFMIK).Utilizing contextual knowledge to guide the forecasting process,this method can cover the influence of those factors that cannot be explicitly included in the forecasting model,and thus it can decrease the forecast error from stochastic events to some extent.Through a container throughput forecast case,this paper compares the performance of CFMIK,AFTER(a combined forecasting method) and 3 single models(ARIMA,BP-ANN, Exponential Smoothing).The results show that the performance of CFMIK is better than that of the remaining ones.