A grey spatiotemporal incidence model with application to factors causing air pollution

Jing Sun; Yaoguo Dang; Xiaoyue Zhu; Junjie Wang; Zhongju Shang

doi:10.1016/j.scitotenv.2020.143576

A grey spatiotemporal incidence model with application to factors causing air pollution

Sci Total Environ. 2021 Mar 10:759:143576. doi: 10.1016/j.scitotenv.2020.143576. Epub 2020 Nov 17.

Authors

Jing Sun¹, Yaoguo Dang¹, Xiaoyue Zhu², Junjie Wang³, Zhongju Shang¹

Affiliations

¹ College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211100, China.
² College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211100, China; Department of Systems Design Engineering, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada.
³ College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211100, China. Electronic address: wangjj@nuaa.edu.cn.

PMID: 33272599
DOI: 10.1016/j.scitotenv.2020.143576

Abstract

The factors causing air pollution in China has caused extensive concern, but there are still many problems in the grey incidence model of identifying air pollution factors. The results produced by the existing grey incidence models are not stable when the order of objects in a given panel data is changed. In order to improve the reliability and uniformity of the grey incidence model, a new grey incidence model, called the grey spatiotemporal incidence model, abbreviated as the GSTI model, is designed in this paper. In the proposed model, the spatiotemporal data which can represent the spatial relationship among different objects rather than the three-dimensional panel data are defined. In addition, the new model includes two procedures. Firstly, the trend coefficient is used to measure the positive and negative connections between two data sequences. Secondly, the measurement coefficient is utilized to calculate the size of grey incidence degree. Subsequently, five properties of the GSTI model are discussed. To demonstrate its practicability and compatibility, the novel model is utilized to identify south Jiangsu province's main factors causing air pollution according to monthly data for 2018. The abundant comparison shows the applicability and superiority of the model in the identification of air pollution factors and the construction of grey incidence model.

Keywords: Air pollution; Grey incidence model; Spatial dimension; Spatiotemporal data; Time dimension.