IMPLEMENTATION OF THE GENERALIZED SPACE TIME AUTOREGRESSIVE (GSTAR) MODEL IN THE CASE OF THE SPREAD OF CORONAVIRUS IN THE DISTRICT CITY OF NORTH SUMATRA

Abstract

Corona Virus Disease 2019 (COVID-19) is a new virus that can be transmitted and the worst impact is death. Covid-19 first appeared in Wuhan, China and eventually spread throughout the world, one of which was North Sumatra Province. The spread of Covid-19 cases was quite rapid, until finally the World Health Organization (WHO) declared the Covid-19 case a pandemic. Based on the conditions that occurred, this final project discusses the prediction of positive cases of Covid-19 in five locations in North Sumatra using the Generalized Space Time Autoregressive (GSTAR) model. Considering that Covid-19 spreads very easily, it does not only depend on time but also the proximity between locations, so the GSTAR model is quite good to use in predicting it, assuming the parameters between locations are heterogeneous. The estimation used is OLS with inverse distance weight. This study aims to determine the best GSTAR model and forecast positive cases of Covid-19 at five locations in North Sumatra. The results show that the best GSTAR model in this study is -OLS with an inverse weight of distance with forecasting results for the next 10 days in May 2022.

Keywords

• gstar
• covid-19
• inverse distance location location weighting matrix
• ols