A Grey Incidence Model for Panel Data Based on the Curvature of Discrete Surface

To determine the relationship between panel data, a grey incidence analysis model based on the curvature of the discrete surface, namely the grey discrete curvature incidence model (GDCI), is proposed in this paper. Firstly, panel data are projected as discrete triangular surfaces. Secondly, based on the Mean curvature and the Gauss curvature of the discrete surface, the coefficient formulae of grey incidence of the Mean curvature and the Gauss curvature are respectively constructed. Then, two grey incidence models based on the Mean curvature and the Gauss curvature are established, respectively. Subsequently, a grey incidence model is proposed based on the curvature of discrete surfaces for panel data. The properties of the proposed model, e.g., normality, symmetry, similarity, and invariance to translation, are also discussed. Finally, both a numerical example and a practical example are given to illustrate the effectiveness and rationality of the proposed model. These examples also indicate that the proposed model can reflect the relationship between the panel data.