LSTM and ARIMA for Forecasting COVID-19 Positive and Mortality Cases in DKI Jakarta and West Java

ABSTRACT

The spread of COVID-19 in Indonesia is still classified as a pandemic until October 31, 2022. Even though the endemic has been enforced in several nations worldwide. However, the fact that people's mobility is increasing means that this condition can increase the number of new cases of COVID-19. The Indonesian government remains vigilant about any decisions that will be taken to maintain the stability of the country's health sector, economy, and population mobility. First, The purpose of this our research is to forecast of daily positive confirmed and daily mortality for the next 13 days using COVID-19 epidemiological data in Indonesia, i.e. DKI Jakarta and West Java. Second, the forecasting model uses a deep learning approach, i.e. LSTM and ARIMA. furthermore, The LSTM method and ARIMA modeling results are compared based on their respective to regions. Finally, The LSTM method has good model performance and the ability to forecast COVID-19 cases based on RMSE and MAPE. © 2022 IEEE.