Forecasting for a Data-Driven Policy Using Time Series Methods in Handling COVID-19 Pandemic in Jakarta

The coronavirus diseases 2019 or COVID-19 has spread and infected millions of people around the world. The on-going COVID-19 pandemic has taken an unprecedented toll on residents, business, commerce, and activity in many cities, including Jakarta, where there have been more than twelve thousand confirmed cases as of July 2020. The details of how COVID-19 spreads in Jakarta are still complicated and not completely understood because the number of infections is large and continues to climb. This paper conducts a quantitative analysis of the COVID-19 pandemic spreading using Jakarta as a case study for the evaluation and decision-making process. In this paper, time series models such as the Holt's exponential smoothing and Auto-Regressive Integrated Moving Average (ARIMA) were used to forecast the number of COVID-19 cases in Jakarta between March 1 and July 6. Recently, data exploration and comparative analysis of time series models have been conducted to determine the optimal models for forecasting COVID-19 confirmed cases. The result shows that ARIMA has the highest R-Squared (R2), and lowest (Mean Squared Error) MSE and Root Mean Squared Error (RMSE) is the best model to forecast the upcoming number of infected cases of COVID-19 in Jakarta. Such a model shows promising results and fitting predictions in supporting data-driven policy in public health and epidemiology.