In order to enhance forecasting precision and easy to modeling, particular analyzed application rule of all-possible-regression analyse, this method was used in managing in the original data, select right import variable, then eatabish forcast model combined with back propagation,, and get preferable forecasting effect.
从提高预测精度和方便建模的角度出发,分析了所有可能回归分析法的应用准则,探讨了应用所有可能回归分析法对初始样本数据处理,选取合适的输入变量,结合BP神经网络建立预测模型的过程,并且取得较好的预测效果。
The problem of predicting variables of ordinal scale is referred to as ordinal regression and is complementary to the standard machine learning tasks of classification and metric regression.
根据专家决策和结构风险最小化原则,构建了一种新的不确定性有序 SVR风险预警模型,可以解决训练点带有不确定性的有序回归问题。
