Clustering the countries for quantifying the status of Covid-19 through time series analysis

Abstract

Purpose This study aims to use financial stability and health facilities of countries, to cluster them for making a more consensus environment for manifesting the status of Covid-19 in a justifiable manner. The scarcity of the categorisation of the countries of the world in a common platform, and the requirement of manifesting the pandemic status such as Covid-19 in a justifiable manner create the demanding requirement. This study mainly focusses on assisting to generate a liable manifesto to criticise the span of viral infection of the severe acute respiratory syndrome coronavirus-2 over the globe. Design/methodology/approach Data for this study has been gathered from official websites of the World Bank, and the world in data. The Louvain clustering method has been used to cluster the countries based on their financial strength and health facilities. The resulted clusters are visualised using Silhouette plots. The anomalies of the clusters had been used to quantify the pandemic situation. The status of Covid-19 has been manifested with the time series analysis through python programming. Findings The countries of the world have been clustered into seven, where developed countries divided into three clusters and the countries with transition economies and developing clustered together into four clusters. The time series analysis of recognised anomalies of the clusters assist to monitor the government responses and analyse the efficiency of used safety measures against the pandemic. Originality/value This study’s resulted clusters are highly valuable as a division of countries of the whole world for evaluating the health systems and for the regional levels. Further, the results of time series analysis are beneficial in monitoring the government responses and analysing the efficiency of used safety measures against the pandemic.

Sociodemographic Factors Associated with Covid-19 Mortality in Brazilian Municipalities: an approach supported by clustering

“…In the literature, some studies identified patterns of the Covid-19 impact in different countries [4][5][6][7][8]. Other works have used clusters to analyze the Covid-19 impact on counties across the United States [9][10][11].…”