Forecasting COVID-19 cases for Top-3 countries of Southeast Asian Nation

Abstract

Several countries continue controlling the spread of the corona virus to decrease the number of new COVID-19 cases. Currently some Southeast Asian countries require an estimate of the num-ber of daily new COVID-19 cases of in the future in order to reopen or consider lifting strict pre-vention policies. This study applies ARIMA and SARIMA forecasting models to predict the de-cline in the number of new cases in three Southeast Asian countries. The first modelling is carried out using the ARIMA model with optimized model parameters based on the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) analysis. Then, the Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and Mean Absolute Percent Error (MAPE) are evaluated is applied a criterion to select the best model. The best ARIMA and SARIMA models are selected manually and they are used to predict the number of new cases in three Southeast Asian coun-tries. It is expected that the number of new cases in these countries will experience a significant decline in the next month from September 2021. The prediction of SARIMA model indicates a better result than the ARIMA model which confirms the existence of a season in COVID-19 data.

Keywords

• covid-19
• arima
• sarima
• prediction