Estimating High Dimensional Spatial SIRD Parameters Using a Heuristic Algorithm, Study Case: Forecasting of COVID-19 Spread in Indonesia

The coronavirus COVID-19 is a pandemic spreading around the world. Indonesia is one of the countries infected. Recently the spread of COVID-19 in Indonesia has been increasing exponentially since the Indonesian government officially announced the first case in April 2020. A spatial Susceptible-Infected-Recovered-Dead (SIRD) model is used to describe the COVID-19 spreading in 34 provinces in Indonesia. The proposed model captures the infection, recovery, death, and mobility rate in each Indonesian province which gives a high dimensional parameterized model. The spatial factor is manifested in the form of origin-destination of public transportation between the provinces. To estimate the high dimensional parameters of the SIRD model, we propose a heuristic algorithm. The daily reported COVID-19 cases in each province were used in the model. The results show that the proposed model can capture the high dimensional parameters and is suitable for predicting the short-term corona-virus behavior in Indonesia, particularly in estimating the number of infected cases.Funding: This research was supported by the Indonesian Ministry of Research, Technology, and Higher Education, through the scheme of Doctoral Dissertation Research Grant (PDD 2021).Declaration of Interests: The authors declare no conflict of interest.