Table of Content

29 October 2022, Volume 34 Issue 3

Previous Issue    Next Issue

A Clustering Evaluation Models for Grey Panel Data Based on the Possibility Function

Lirong Sun, Wencheng Li, Jiahui Ma, Danlei Feng

2022, 34(3):  1-20. 

Asbtract ( 144 )  

Related Articles | Metrics

For data with the characteristics of "small sample, poor information," the concept of grey panel data is proposed, and a clustering method with the grey panel data based on the possibility function is developed. First, the application range of the possibility function is expanded, and a possibility function based on the interval grey number is derived. Afterward, aiming at the problem that the existing model cannot be applied to the grey panel data directly, a method for determining cluster weights of the grey panel data is proposed. According to the principle of new information priority, the grey cluster coefficient matrix at different moments is integrated to obtain the comprehensive grey cluster coefficient matrix. Finally, the objects are divided into grey categories under the grey cluster coefficient maximization principle. The proposed method is applied to the air quality assessment of eight major cities in China. Compared with the traditional panel data clustering method, it is found that the proposed method can refine and stratify the quality of the clustering results, which can make the clustering results clearer and easier to understand.

Bayesian Network Model of China’s Financial Risk Under COVID-19 Based on Grey Clustering

Shuli Yan, Jiacheng Feng, Na Zhang, Xiangyan Zeng

2022, 34(3):  21-35. 

Asbtract ( 119 )  

Related Articles | Metrics

The panic caused by COVID-19 and the stagnation of business activities induced the continuous breeding of China’s financial risks. This paper considers the COVID-19 and economic indexes as nodes to establish the Bayesian topology of financial risk. The liquidity, sovereign, and stock market risks are mainly considered to evaluate the financial risk. Based on the risk characteristics, the central interval trapezoidal possibility functions are designed, then the grey clustering model is used to classify the financial risk into four different levels. The possibility distribution of financial risk levels under different COVID-19 index levels is inferenced through the Bayesian network. Finally, each node’s monthly time series data from October 2019 to May 2021 is used to learn by NETICA software, and the conditional probability of each node and the possibility of financial risk are deduced. It is concluded that liquidity risk and sovereign risk are more sensitive to COVID-19, while the stock market risk is not very sensitive to it.

Prediction and Analysis of Cerebrovascular Disease Mortality Based on Grey Model 

Xuejun Shen, Honglin Hu, Chunyi Huang, Hongyue Yan, Shuyi Zhou, Jufang Li, Weiping Li, Xuerui Tan

2022, 34(3):  35-46. 

Asbtract ( 77 )  

Related Articles | Metrics

This paper aims to explore the optimum modelling data for predicting cerebrovascular disease mortality in China by using GM(1,1) model. Also, it will discuss the applicability of grey relational analysis to analyze the related factors of cerebrovascular disease mortality. The subsequence sets with different data lengths were used to establish rolling GM(1,1), and predictive efficiency was judged by prediction error. Grey relational analysis was used to analyze the interrelationships between population, health expenses, environmental pollution factors and mortality. Results show that GM (1,1) grey prediction model can well predict the mortality of cerebrovascular diseases in China, especially when the modelling data is 8, the overall prediction effect is the best. In recent years, the prediction effect of 4 or 5 data modelling is better than that of more data modelling. Grey relational analysis suggests that population aging has a greater impact on the mortality of cerebrovascular diseases than other risk factors. The grey system theory is suitable for the epidemiological study of cerebrovascular diseases.

BOPS Channel Strategies for Manufacturers and Online Retailers Under Omnichannel Operations Using Grey Game 

Yinhai Shen, Qing Zhang

2022, 34(3):  47-65. 

Asbtract ( 99 )  

Related Articles | Metrics

Both traditional manufacturers and online retailers have been implementing omnichannel strategies such as Buy-Online-and-Pick-up-in-Store (BOPS). By introducing interval grey numbers to represent consumer preference, which depicts uncertain consumer valuation, and the wholesale price and selling price constitute the profit function, we develop a basic dual-channel and three omnichannel grey models. Compared with the traditional online or offline dual-channel supply chain, the construction of the omnichannel model considers agency selling agreements and spillover effects as well. Four propositions have been proposed and proven through the analysis of the position degree analysis of the grey game model. Equilibrium analysis indicates that there is no option of [BOPS, No BOPS] in the equilibrium strategy of the grey game. Finally, a case study and numerical simulation are given to verify the model and reveal the relationship between the profit function and BOPS channel agency fees when reaching an equilibrium strategy.

A Bibliometric Analysis on Grey System Theory and Its Application in 1982–2021

Yuying Yang, Naiming Xie, Bin Liu, Caorui Liu

2022, 34(3):  66-80. 

Asbtract ( 78 )  

Related Articles | Metrics

This paper analyses the development level and trend of grey system theory (GST) over the past 40 years based on bibliometrics. Literature was searched using the Web of Science (WoS) databases. Literature analysis was carried out from eight aspects: paper publication, main journals publishing GST and its application, 20 highly cited articles, hot research topics, high-output scholars and their cooperation networks, the geographical distribution of scholars, funding agencies, and a comparison between papers obtained from the China National Knowledge Infrastructure (CNKI) and WoS databases. Bibliometric analysis showed that while GST had developed slowly over the initial 20 years, it experienced a period of high-speed development over the last 20 years. Two journals—the Journal of Grey System and Grey Systems: Theory and Application—play key roles in promoting the international development of GST. Professor Sifeng Liu and Nanjing University of Aeronautics and Astronautics have been the most influential scholar and research center in GST, respectively. GST has attracted many scholars worldwide, and the number of papers published in international journals is increasing. A comparison of publications related to GST between CNKI and WoS databases showed that more efforts are required for GST to become more international.

A Watermarking Algorithm Using QR Code and Grey Relational Analysis in DCT Domain

Qiuping Wang, Fan Tang, Fang Dai, Xiaofeng Wang

2022, 34(3):  81-94. 

Asbtract ( 81 )  

Related Articles | Metrics

An image watermarking algorithm in a discrete cosine transform domain based on QR code and grey relational analysis is presented in this paper. Literal information is encoded with a QR code as the digital watermark, and the watermark image is scrambled before embedding. The host image is divided into some non-overlapping blocks, and each block is processed with a two-dimensional discrete cosine transform. According to the correlation among image pixels, non-smooth regions of the host image are selected to embed watermark information by computing a double-direction grey correlation degree. Watermark embedding is selected from two schemes. One is based on the singular value decomposition (SVD), and the other is based on the orthogonal triangle decomposition (QR decomposition). Contrast experiment results show that the watermarking algorithm named QR code-GRA-SVD using the watermark embedding scheme based on singular value decomposition not only ensures the imperceptibility of the watermarking algorithm but also owns strong robustness. 

Construction and Application of a Time-Delayed Grey Bernoulli Model With Dummy Variables

Xue Bai, Shi Yao, Ye Li

2022, 34(3):  95-114. 

Asbtract ( 63 )  

Related Articles | Metrics

To address the issue that the traditional grey model is difficult to predict nonlinear data series with multiple mutations caused by policy evolution, a time-delayed grey Bernoulli model with dummy variables (abbreviated as DTD-NGBM(1,1)) is proposed by introducing policy effects as dummy variables. Next, the solution of the time response function of DTD-NGBM(1,1) is discussed, and the optimal values of the time-delayed term and the nonlinear parameter are determined using the debugging method and the genetic algorithm, respectively. Finally, the validity and superiority of the proposed model are verified by taking solar power generation volume forecasting in China and the U.S. as examples. The results show that the proposed model can accurately describe the trend of the data series under the influence of dummy variables.

A Review of Grey Target Decision Model

Sandang Guo, Qian Li, Yaqian Jing, Liuzhen Guan

2022, 34(3):  115-134. 

Asbtract ( 79 )  

Related Articles | Metrics

An increasing amount of papers described different ways to obtain the ideal scheme by grey target decision model but seldomly set out the relative benefits of each approach so that the choice of the approach seems arbitrary. Therefore, this study discusses various grey target decision techniques and guides researchers in choosing suitable techniques for different decision models. This paper reviews the literature about grey target decisions published from 2010 to 2020, particularly methodological innovation. These techniques are categorized by the five aspects of developing a grey target decision model: (i) main types of data, (ii) manipulation of data, (iii) determination of criteria weights, (iv) calculation of bull’s-eye-distance, and (v) hybrid models. These techniques are discussed in terms of their underlying principles, complexity, strengths, and weakness. Summary tables and specification charts are given to guide the selection of suitable techniques. 

A Novel Grey Seasonal Prediction Model for Container Throughput Forecasting

Yichung Hu, Geng Wu, Shuju Tsao

2022, 34(3):  135-147. 

Asbtract ( 76 )  

Related Articles | Metrics

Containerization is regarded as an important driver of globalization and international trade, and it also drives the development of global ports. Seasonal container throughput prediction is crucial for planning and operation by port authorities, and for the strategies formulated by logistics companies. To accurately predict the seasonal fluctuations in port container throughput, we propose a novel grey seasonal model called, FNDGSM(1,1). The proposed model involves time item, cycle Hausdorff fractional accumulating generation, and seasonal dummy variables. The particle swarm optimization algorithm is used to obtain the optimized parameters. Experimental results demonstrate that the proposed seasonal grey prediction model performs significantly better than other prediction models with quarterly container throughput data.

Forecasting Housing Prices in China’s First-Tier Cities Using ARIMA and Grey BR-AGM(1, 1)

Zhongqin Wen, Yichung Hu, Shuhen Chiang

2022, 34(3):  148-173. 

Asbtract ( 104 )  

Related Articles | Metrics

Housing prices in China have grown rapidly and dramatically over the past two decades; at the same time, the housing sector has contributed greatly to China’s economy. Thus, the importance of exploring China’s housing question cannot be overemphasized. To better understand the dynamics of housing prices in China, we try to forecast housing prices in China’s first-tier cities: Beijing, Shanghai, Guangzhou, and Shenzhen, by means of rolling ARIMA models and Grey BR-AGM (1,1) model in order to compare their forecasting performances. The empirical results demonstrate that Grey BR-AGM (1,1) model outperforms other models with a quicker reaction to external policy shocks.