Temporal and spatial evolution of the distribution related to the number of COVID-19 pandemic

TL;DR

This study analyzes the temporal and spatial evolution of COVID-19 distributions across countries, revealing power-law and stretched exponential behaviors that reflect the intrinsic dynamics of virus spread over two years.

Contribution

It provides a systematic analysis of COVID-19 distribution patterns, identifying key statistical behaviors and their evolution, which aids in understanding and modeling the pandemic.

Findings

Distributions follow power-law in early stages

Distributions follow stretched exponential in later stages

Distribution behaviors reflect underlying virus spreading dynamics

Abstract

This work systematically conducts a data analysis based on the numbers of both cumulative and daily confirmed COVID-19 cases and deaths in a time span through April 2020 to June 2022 for over 200 countries around the world. Such research feature aims to reveal the temporal and spatial evolution of the country-level distribution observed in COVID-19 pandemic, and obtains some interesting results as follows. (1) The distributions of the numbers for cumulative confirmed cases and deaths obey power-law in early stages of COVID-19 and stretched exponential function in subsequent course. (2) The distributions of the numbers for daily confirmed cases and deaths obey power-law in early and late stages of COVID-19 and stretched exponential function in middle stages. The crossover region between power-law and stretched exponential behaviour seems to depend on the evolution of "infection" event and "death" event. Such observation implies a kind of important symmetry related to the dynamics process of COVID-19 spreading. (3) The distributions of the normalized numbers for each metric show a temporal scaling behaviour in 2-year period, and are well described by stretched exponential function. The observation of power-law and stretched exponential behaviour in such country-level distributions suggests underlying intrinsic dynamics of a virus spreading process in human interconnected society. And thus it is important for understanding and mathematically modeling the COVID-19 pandemic.