灰色预测中缓冲算子的组合性质及应用

通过研究同性缓冲算子作用样本数据序列所得灰色预测模型的预测精度问题, 发现同类缓冲算子对于相同样本数据序列的预测值一致, 对于具体预测问题, 某个缓冲算子在此类问题中有效, 用到另一类问题上可能失效. 为此, 提出一种同性缓冲算子的组合方法, 通过对同性缓冲算子的有效组合可以得到一种新的缓冲算子, 该缓冲算子不但能够提高模型的预测精度, 而且可以扩大缓冲算子的适用范围. 实例验证表明, 所提出的组合方法是有价值的.

Properties and applications of combined buffer operators in grey prediction

By studying the prediction accuracy of the grey prediction model through the same type of buffer operator, it is found find that the predict result of the same buffer operator is very different in the same sample data sequence. For the specific prediction problem, a buffer operator is effective in this kind of problem, but ineffctive in anothers. Therefore, a new method is proposed to improve the prediction accuracy of the grey model, and the application range of the buffer operator can be expanded by using the combination method of the same type of buffer operator. Examples are given to illustrate that the proposed method is valuable.