Prediksi Jumlah Pasien Positif Covid-19 Di Indonesia Menggunakan Model Berbasis Spasio Temporal GSTAR Orde Satu

Abstract

Coronavirus Disease 2019 (COVID-19) is a pandemic disease that has not been previously identified in humans. The virus that causes COVID-19 is called Sars-CoV-2. And this corona virus is zoonotic (transmitted between animals and humans). The spread of COVID-19 can be through droplets (small particles) when someone talks or sneezes, air, and contaminated surfaces. So that the main factors causing the increase in COVID-19 include increased movement, activity, and interaction of the population, such as activities in public transportation and the workplace, then the new variant factor of COVID-19 and the linkage in the previous time. The process of spreading from one location to another (transmission) involves a spatial process. The COVID-19 time series data can be modeled with the spatio-temporal-based GSTAR model on 3 islands in Indonesia, namely Java Island and Sulawesi Island. The weight used in this study is the inverse distance weight with the appropriate GSTAR model being GSTAR (1,1). The forecast level of the GSTAR model for all testing and training data with Inverse Distance weights which has the smallest RMSE is in the GSTAR model for Location Java, which is 0.40255. Meanwhile, the forecast for the GSTAR model which has the largest RMSE value is Sulawesi Island, which is 1.616303.