论文部分内容阅读
传统的组合预测方法中,各单项预测模型的加权系数一般是固定不变的。而事实上,任何一种预测模型的预测精度都会随着预测时点的不同而发生改变。因此变权组合预测模型的研究更有意义和价值。本文在预测有效度和IOWA算子等概念的基础上,建立了以预测有效度和误差平方和为准则的组合预测模型,并给出了模型在实际预测中的应用,结果表明了新模型是一种优性组合预测。
In the traditional combined forecasting method, the weight coefficients of each individual forecasting model are generally fixed. In fact, the prediction accuracy of any one of the prediction models will change with different points in time. Therefore, the study of variable weight portfolio forecasting model is more meaningful and valuable. Based on the concepts of predictive validity and IOWA operator, this paper establishes a combined forecasting model based on predictive validity and square sum of errors, and gives the application of the model in the actual forecast. The results show that the new model is An Optimal Combination Forecasting.