灰色关联模型从统计序列几何形状的相似程序来判断其紧密程度,传统的灰色关联模型从统计序列的面积、斜率、变化速率等角度来度算其关联程度,这些关联模型受到统计序列横纵坐标(周期和振幅)的双重影响,这在分析统计序列间的单个因素(本文研究仅受周期影响)间的关系时会产生其它因素的干扰。针对这一缺陷,作者根据两条曲线的曲折情况,通过计算各折线的斜率判断曲线的波动周期,提出基于斜率判断的灰色周期关联度。该方法计算简单、判断直观,既可以有效解决上述文献中的问题,又具有传统灰色关联度良好的性质,并通过实例验证该方法的实用性。
Grey incidence model determines the closeness degree from the closeness and similarity of statistical sequence's geometrical form. Traditional grey incidence model calculates statistical sequence's correlation degree from the perspectives of its acreage, slope, changing velocity, etc. These incidence models have dual influences exerted by horizontal and vertical coordinates. Subsequently, inference of other factors occurs, when we analyze the relationship between statistical sequence's individual factors. To overcome this limitation, the paper proposes the degree of grey periodic incidence judged by slope, according to the winding situations of two curves, and judging wave period of the curve by calculating the slope of every polygonal line. The calculation is simple and intuitive on judgment by the method. It can not only solve effectively the problem of above reference, but also have good quality of traditional gray periodic incidence. At last, the paper validates its practicability with a practical case.