论文部分内容阅读
建立一个以预测误差平方和达到最小为准则的正权重组合预测模型。以成都私家车数量预测为例,分别采用灰色预测模型、指数模型、一次函数模型、二次函数模型、三次函数模型做单项预测。通过组合预测,得到更高精度的预测结果。以最优组合预测模型预测成都市在2009年的私家车数量,并与实际值进行比较,对比分析计算误差。同时也运用组合预测法对2010年成都私家车数量进行了预测。
A positive weighted portfolio forecasting model based on the minimization of the square of the forecasting error is established. Taking the prediction of the number of private cars in Chengdu as an example, we use the gray prediction model, the exponential model, the one-time function model, the quadratic function model and the cubic function model to make the single prediction. By combining predictions, more accurate predictions are obtained. The optimal combined forecasting model is used to predict the number of private vehicles in Chengdu in 2009 and compare with the actual value to compare and analyze the calculation error. At the same time, the forecast of the number of private cars in Chengdu in 2010 is also predicted by the combination forecasting method.