Журнал инфекционных заболеваний и терапии

Абстрактный

Linear Regression Analysis of COVID-19 Time-Series Data using the Gumbel Distribution

Hiroshi Furutani, Tomoyuki Hiroyasu

This study uses the Gumbel distribution to model and analyze the daily number of COVID-19 deaths in 8 European and North American countries, as well as in the 7 NHS regions of England, during the first wave of the COVID-19 outbreak. Linear regression is used for parameter estimation and data fitting. The analysis focuses on the height and position of the peak as indicators of the efectiveness of the algorithm. The results of the proposed approach show that the Gumbel model reasonably reproduces the time-series data of COVID-19 deaths in many regions. The advantage of the proposed method is its simplicity and straightforwardness, which allow us to obtain preliminary results for an intuitive image of trends without the need for a sophisticated mathematical framework.