Forecasting Number of COVID-19 Cases in Indonesia with ARIMA and ARIMAX Models

Aji, B. S.; Indwiarti, Rohmawati A. A.

During the pandemic COVID-19, Indonesia has a significant number of positive cases among countries in Asia. In early December 2020, the death rate in Indonesia had been reached more than 3%. Meanwhile, the daily number of positive is also continued to increase, it happens due to lack of anticipation rules made by local authorities and central government. Thus, the preventive step such forecasting becomes a major issue in the area of science and technology, to make all stakeholders well-prepared against this pandemic. This paper provides the performance of The Autoregressive Integrated Moving Average (ARIMA) to forecast several COVID-19 and also examines Auto Regressive Integrated Moving Average with exogenous variables (ARIMAX) model by considering Google Trends as an external variable. We consider a daily dataset from the official website of Jakarta's COVID-19 and the Google Trends data based on certain queries as external variables on March 1-November 25, 2020. According to ARIMA and ARIMAX models, we have ARIMAX model with Google Trends improving ARIMA's performance by reducing the MAPE by 0.8%. © 2021 IEEE.