灰色关联度模型正负性问题的研究及其改进

灰色关联分析是通过对序列几何关系的度量来比较系统中多因素关系的紧密程度,其紧密程度可由序列曲线几何形状的接近(或相近)程度来度量。因此可以利用参考序列和比较序列各对应时段斜率的比值来反映序列之间的紧密程度。首先定义了广义变异系数,并提出一种新的关联度计算方法,该关联度能够反映序列的正相关与负相关关系,并且只与序列的形状有关,而与它们在空间的相对位置无关。此方法充分利用了序列各时点的信息及正、负相关性。最后通过算例验证了它的合理性与实用性。

Study and improvement on properties of positive and negative of grey incidence degree model

Grey incidence analysis is a multiple factors method that the closeness of relationship of the multiple factors is judged based on the similarity level of the geometric patterns of sequence curves.Therefore,the slope ratio of reference sequences and relevant sequences on every corresponding interval is used to reflect the closeness of the sequences.The article defines the generalized variation coefficient and puts forward a new kind of grey incidence model.This kind of model reflects the positive and negative relation of grey incidence degree to some extent.It is only relate to the geometric shape of sequences and has nothing to do with the spatial position of theirs.The method takes full advantage of every intervals' information and the positive and negative relations.At last,an example is given to prove the rationality and practicability.