Gray clustering model based on the degree of dynamic weighted incidence for panel data and its application
Grey Systems: Theory and Application
ISSN: 2043-9377
Article publication date: 10 October 2020
Issue publication date: 16 October 2020
Abstract
Purpose
The paper aims to propose a clustering model for panel data. More specifically, the paper aims to construct a gray incidence model for panel data to solve the classification with multi-factors and multi-attributes.
Design/methodology/approach
The paper opted for a clustering theory study using gray incidence theory based on dynamic weighted function. The paper presents an example to verify the rationality of the new model, which suggests that the new model can reflect the incidence degree of panel data.
Findings
The paper provides a new gray incidence model based on a dynamic weighted function that can amplify the characteristics of the sample to some extent. The properties of the new incidence model, such as normalization, symmetry and nearness, are all satisfied. The paper also shows that the new incidence model performs very well on cluster discrimination.
Originality/value
The new model in this paper has supplemented and improved the gray incidence analysis theory for panel data.
Keywords
Acknowledgements
This work was supported by University of Jinan Science and Technology Development Fund (XKY1614) and A Project of Shandong Province Higher Educational Science and Technology Program (J18KA249).
Citation
Wu, H. and Qu, Z. (2020), "Gray clustering model based on the degree of dynamic weighted incidence for panel data and its application", Grey Systems: Theory and Application, Vol. 10 No. 4, pp. 413-423. https://doi.org/10.1108/GS-09-2019-0040
Publisher
:Emerald Publishing Limited
Copyright © 2019, Emerald Publishing Limited