Application of combined forecasting method to prediction of demand for the special purpose vehicle in China

This paper predicts the demand of the special purpose vehicle for the next few years based on the actual data of special purpose vehicle production in China since 1998. In order to improve the accuracy of forecasting, it establishes two kinds of combination forecasting model through the introduction of weight, according to the different methods to handle error. The composite models integrate the advantages of three single forecasting methods: the exponential smoothing, growth function and the GM (1, 1) model. Optimal combination forecasting model, which is determined by minimizing the sum of squared errors, makes the forecast more accurate. The predictive result will play an important guiding role in coordinating the supply and demand of special purpose vehicle and avoiding the mistakes in investment decision-making caused by out of control.