Research on extended cluster of grey incidence and its application

This paper studies extended clusters of grey incidence for indexes in panel data. First, it discusses the expression of a panel data by matrix sequence, and describes its geometric characteristic through a cluster of curved surfaces. Thereafter, in three dimensional space, the paper analyzes the geometric similarity of indexes in panel data. According to the principle of grey incidence analysis, it proposes an absolute degree of grey incidence for matrixes, which maintains the consistent form with the classic one. Furthermore, the formula of zero starting point operator and parameters are redefined. On this basis, the paper constructs the extended incidence matrix of indexes and discusses the cluster analysis process of panel data. Finally, an example is presented to illustrate the proposed method. The results prove the method is effective.