SCM and SARIMA on Forecasting COVID -19 Outbreak in Italy

The COVID -19 originated from China in the year 2019 and grabbed more than 173,321, 279 human lives with a death total of more than 3,727,177 globally. Italy had been experienced with worst and the first European country severely hit by the COVID-19. At present daily infected cases shows a declining trend in Italy. Yet, the future is uncertain, it could rise again. The dynamic changes of the outbreak should be captured more often. Hence, the study has designed to forecast the daily infected cases of COVID -19 in Italy. The daily infected cases of COVID-19 of Italy for the period of 22nd January 2020 to 3rd June 2021 were obtained from the World Health Organization (WHO) database. The pattern recognition of the daily infected cases examined by time series plots and Auto Correlation Function (ACF). Based on the pattern the Auto-Regressive Integrated Moving Average (SARIMA) and the Sama Circular Model (SCM) were tested to forecast the daily infected cases in Italy. The model assumptions were tested by using the Anderson Darling test, ACF, and Ljung-Box Q (LBQ)-test. The forecasting ability of the models was assessed by three measurements of errors; Mean Absolute Percentage Error (MAPE), Mean Square Error (MSE), and Mean Absolute Deviation (MAD) in both model fitting and verification processes. The results of the study revealed that the SARIMA and SCM were satisfied all validation criteria and the performance of the models was extremely high. It had been concluded that the SARIMA and SCM are suitable to forecast daily infected cases in Italy. It is strongly recommended to identify the hidden seasonal behaviors of other countries furthermore.