Grey incidence model of B-mode based on panel data and its application

Based on the basic idea of the grey relational analysis of B-mode, aiming at the defects of the association order inconsistency caused by the change of object arrangement sequence in the grey relational model of the existing panel data and the problem of distortion caused by not fully considering the change rate between object dimension sequence and the mean for different objects at the same time, the gray incidence model of B-mode based on panel data is constructed from two aspects of time dimension and object dimension. In the time dimension, the horizontal correlation degree is obtained by introducing the overall displacement difference, the first order slope difference and the two order slope difference between each index. In the object dimension, the vertical correlation degree is described by the ratio of each point and the mean of the different objects at the same time. And then, the grey incidence model of B-mode based on panel data is constructed by calculating the weighted average of the two correlation degrees. The properties of normalization and multiple transformation isotonicity of the model are discussed. The example shows that the model is simple and effective, and is not affected by the order of objects. Finally, the drought hazard index of five cities in Northern Henan Plain is used as the characteristic index sequence to clarify the relationship between the drought risk index and its 12 influencing factors, which provides a theoretical basis for the drought risk assessment and regulation. © 2020, Editorial Office of Control and Decision. All right reserved.