Modeling Daily Infected Cases of COVID -19 in Ukraine

COVID-19 is the worst pandemic in the 21st Century after the Swine Flu in 2009-2010. The World Health Organization (WHO) declared COVID-19 to be a pandemic when it was spreading very fast over a wide area. Ukraine has reported the 8th highest European country and 3rd highest Eastern European country grabbed by COVID-19. Ukraine exceeds 3.5 million cases since 22nd January 2020. At present, the daily infected cases show a decreasing trend. This study is designed to model daily infected cases in Ukraine. The daily infected cases in Ukraine for the period of 22nd January 2020 to 12th December 2021 were obtained from the Humanitarian Data Exchange (HDX). The behavior of the data series is recognized by time series plots and Auto Correlation  Function (ACF). It was found that the series follow irregular wave-like patterns with increasing amplitudes as well as decreasing amplitudes. The Sama Circular Model (SCM), Seasonal Auto-Regressive Integrated Moving Average (SARIMA), and Holt's Winters additive and multiplicative models were tested to forecast the infected cases. The relative and the absolute measurements of errors were used to assess the ability of the models. The results of the study revealed that Holt's Winters models and the SCM are suitable for the purpose. It also found that the SCM outperformed Holt's Winters models. However, the literature revealed that the Damped Circular Model (DCM) and Forced Circular Model (FCM) would be more suitable to capture the behavior of series with decreasing amplitude and increasing amplitude respectively. Therefore it is recommended to test the DCM and FCM to forecast the daily infected cases of Ukraine and other countries with similar patterns.