Modeling Wave-Like Patterns of Daily Infected Cases of COVID -19 in the UK

The World Health Organization (WHO) declared COVID-19 to be a pandemic when it became clear that the illness was severe and that it was spreading very fast over a wide area. As a result, there were more than 252 million infected cases and 5million death total reported globally after one year. The United Kingdom (UK) is the highly affected European country during the end of the year 2021. Hence, the study is designed to model daily infected cases in the UK. The daily infected cases of the UK for the period of 22nd January 2020 to 17th October 2021 were obtained from the Humanitarian Data Exchange (HDX). The behavior of the daily infected cases is recognized by time series plots and Auto Correlation Function (ACF). The Sama Circular Model (SCM) and Seasonal Auto-Regressive Integrated Moving Average (SARIMA) were tested to forecast the daily infected cases. ACF, Anderson Darling test and Ljung-Box Q (LBQ)-test were used to validate the model. The performance of the model was assessed by the relative and absolute measurements of errors. The results of the study revealed that the SCM is outperformed SARIMA. The SCM has captured 4 repeating behaviors whereas SARIMA has captured only 1 behavior. Measurements of errors were very low in SCM compared with SARIMA in forecasting. Four repeating behaviors of daily cases in the UK are identified. It is recommended to find the underlying factors for such behaviors. Also, it is worth conducting similar studies for other countries as well