论文部分内容阅读
多指标面板数据包含研究对象的诸多信息,但复杂的数据结构也给分析与测度截面个体之间的相似性带来了一定的困难,针对这一问题,探讨了多指标面板数据的矩阵表示,分析了多指标面板数据的时空特征,基于多指标面板数据的时空特征构建了可用于多指标面板数据截面相似性测度的灰色矩阵关联分析模型,该模型将能够表征面板数据时空特征的“水平”距离、“增量”距离、“变异”距离引入灰色关联度计算模型,并将灰色关联度计算由一般向量空间拓展到矩阵空间。并以我国区域道路交通事故成因分析为例,对我国区域道路交通事故的成因进行了分析,检验了模型的有效性。
The multi-index panel data contains many informations of the research object, but the complex data structure also brings some difficulties to the similarity between the individuals of the cross-section analysis and measurement. In view of this problem, the matrix representation of multi-index panel data is discussed, Based on the spatio-temporal features of multi-index panel data, a gray matrix correlation analysis model that can be used to measure similarity of multi-index panel data is constructed. This model will be able to characterize the “ ”Distance “ incremental ”distance, “ variation ”distance into the gray incidence calculation model, and expand the gray incidence calculation from the general vector space to the matrix space. Taking the analysis of the cause of regional road traffic accidents in our country as an example, this paper analyzes the causes of road accidents in our country and tests the validity of the model.