Forecasting Number of New Cases Daily COVID-19 in Central Java Province Using Exponential Smoothing Holt-Winters

Autor: Irandi, Dinda Fitri, Rohmawati, Aniq Atiqi, Gunawan, Putu Harry
Jazyk: angličtina
Rok vydání: 2021
DOI: 10.34818/indojc.2021.6.2.565
Popis: There is hard to mention how long the COVID-19 pandemic will discontinue. There are some factors, including the public’s efforts to slow spread and researchers’ work to observe more about this outbreak. From the beginning of the health crisis, particularly following the announcement of the first positive case In Indonesia due to the COVID-19 on March 2, 2020. Afterwards, the number of daily cases increase simultaneously in other regions in Indonesia until today. Due to the fact that the significant mobility of the people, Central Java has contributed the 3rd rank of potential number of COVID-19 positive cases in Indonesia. This study aims to forecast the number of COVID-19 daily new cases in Central Java to assist the government in preparing the necessary resources and controlling the spread of the COVID-19 virus in Central Java Province. We proposed Exponential Smoothing Holt-Winters with the Additive model with seasonal addition considering trend and seasonal factors. The dataset during March 14 to April 17, 2021, revealed fluctuation of trend and seasonal patterns. Our simulation studies indicate that Exponential Smoothing Holt-Winters provides sharp and well performance for forecasting daily new cases of COVID-19 in Central Java province with MAPE less than 10%.
Indonesia Journal on Computing (Indo-JC), Vol. 6 No. 2 (2021): September, 2021
Databáze: OpenAIRE