Comparisons of COVID-19 dynamics in the different countries of the World using Time-Series clustering

Abstract

In recent months, the world has suffered from the appearance of a new strain of coronavirus, causing the COVID-19 pandemic. There are great scientific efforts to find new treatments and vaccines, at the same time that governments, companies, and individuals have taken a series of actions in response to this pandemic. These efforts seek to decrease the speed of propagation, although with significant social and economic costs. Countries have taken different actions, also with different results. In this article we use non-parametric techniques (HT and MST) with the aim of identifying groups of countries with a similar spread of the coronavirus. The variable of interest is the number of daily infections per country. Results show that there are groups of countries with differentiated contagion dynamics, both in the number of contagions plus at the time of the greatest transmission of the disease. It is concluded that the actions taken by the countries, the speed at which they were taken and the number of tests carried out may explain part of the differences in the dynamics of contagion.

Analysis of communities of countries with similar dynamics of the COVID-19 pandemic evolution

“…The COVID-19 pandemic continues to batter the world, with more than 70 million positive cases and almost two million deaths by the end of December 2020 (see Santiago et al, 2020). This study extends Alvarez et al (2020), in which the authors analyze the evolution of the COVID-19 in a dataset of 191 countries. The paper utilizes a clustering method based on the correlation distance and a Minimal Spanning Tree (MST), from which they obtain a Hierarchical Tree (HT).…”

“…Besides the already-mentioned Alvarez et al (2020), other authors have also employed non-parametric techniques in order to analyze the COVID-19 dynamics. Zarikas et al (2020) analyze the set of 30 countries with the highest number of COVID-19 cases and identify four main clusters during the period between the 22nd of January 2020 and the 4th of April 2020.…”

Investigating the COVID19 Characteristics of the Countries Based on Time Series Clustering

“…They also have investigated the correlation between the population size and spread of COVID19. Alvarez et al [4] have used non-parametric techniques based on correlation distance and Minimal Spanning tree in order to cluster 191 countries in terms of COVID19 dynamics. Hutagalung et al [5] have focused on the grouping of the 11 countries located in Southeast Asia in terms of the number of confirmed cases and the number of deaths observed on the date of April 2020.…”

COVID-19: Worldwide Profiles during the First 250 Days

“…As presented in Table 1, these studies differed in terms of techniques employed as well as time periods, data sources, and regions studied. Three out of these seven studies tried to focus on a more global perspective, studying as many countries as possible [8][9][10] and two considered a temporal analysis, comparing the disease evaluation's similarity over time among different countries [4,11]. However, these latter studies were geographically circumscribed to twelve European countries and to the United States, respectively.…”

“…As presented in Figure 1 and detailed in the previous subsection, since the virus did not affect all countries at the same time, it was also decided to follow the approach of Alvarez et al (2020) to synchronize the scaled data with respect to time. Nonetheless, to have a broader panoramic view and since more data were available, it was decided this time to study a broader period, larger than the first 100 days of the pandemic.…”