最优组合预测方法在家用汽车需求预测中的应用

Application of an Optimal Combined Forecasting Method to Prediction of Demand for Private Cars

为了提高预测的准确性,引入了组合预测模型,将几个单一预测模型有机地结合起来,综合各个预测模型的优点,对未来几年内家用轿车需求进行预测。通过使组合预测误差平方和最小,确定各个单一预测方法的权重系数,得出更为准确的预测结果。计算结果表明该方法具有较好的实用性。

This paper forecasts the demand for private cars in coming years. To improve the accuracy of forecasting, the paper presents a combined forecasting method, which integrates several single forecasting methods. Optimal weights are determined by minimizing the sum of squared errors. The combined forecasting method is rather practicable.