Constructing normal distribution according to media data about COVID-19

Abstract

This paper deals with a hypothesis about the normal distribution of the quantity of victims during the COVID-19 pandemic, and normal distribution curves for the so-called waves identified during this period are constructed. The data on COVID-19 known from the media are taken as a basis, namely the quantity of infected, died and recovered. All necessary information is publicly available. To study the collected information regarding the above indicators, the Sturgess heuristic formula is used to determine the “optimal” number of intervals for partitioning the range of values of the random variable under consideration. For completeness, the study analyzes data on the quantity of infected, died and recovered, collected for each of the three indicators for different countries, taking into account the time frame of specific waves of the pandemic. To confirm the hypothesis of normal distribution, the Pearson goodness-of-fit test, also called the c2 test, is used. In the overwhelming majority of the cases studied, statistical data do not provide grounds to reject the hypothesis about the normal distribution of each of the indicators we are interested in separately, taking into account the time frame of the corresponding waves. The text of the paper shows in detail all steps necessary to verify the validity of the assumption made. All used built-in functions of the Microsoft Office Excel application are indicated, designed to optimize the process of working with large volumes of data and visualize the results for greater clarity. As a main example, statistics for Canada are considered, namely for the first wave of morbidity, which occurred in March – June 2020.