A Multi-Period Curve Fitting Model for Short-Term Prediction of the COVID-19 Spread in the U.S. Metropolitans

• Main Authors: Bilal Majeed (Author), Ang Li (Author), Jiming Peng (Author), Ying Lin (Author)
• Format: Book
• Published: Frontiers Media S.A., 2022-01-01T00:00:00Z.
• Subjects: article
• Online Access: Connect to this object online.
• Notes: 2296-2565; 10.3389/fpubh.2021.809877
• Keywords: health care analysis; coronavirus; multi-period modeling; COVID-19; curve fitting model; Public aspects of medicine; RA1-1270
• Source: Frontiers in Public Health, Vol 9 (2022)

The COVID-19 has wreaked havoc upon the world with over 248 million confirmed cases and a death toll of over 5 million. It is alarming that the United States contributes over 18% of these confirmed cases and 14% of the deaths. Researchers have proposed many forecasting models to predict the spread of COVID-19 at the national, state, and county levels. However, due to the large variety in the mitigation policies adopted by various state and local governments; and unpredictable social events during the pandemic, it is incredibly challenging to develop models that can provide accurate long-term forecasting for disease spread. In this paper, to address such a challenge, we introduce a new multi-period curve fitting model to give a short-term prediction of the COVID-19 spread in Metropolitan Statistical Areas (MSA) within the United States. Since most counties/cities within a single MSA usually adopt similar mitigation strategies, this allows us to substantially diminish the variety in adopted mitigation strategies within an MSA. At the same time, the multi-period framework enables us to incorporate the impact of significant social events and mitigation strategies in the model. We also propose a simple heuristic to estimate the COVID-19 fatality based on our spread prediction. Numerical experiments show that the proposed multi-period curve model achieves reasonably high accuracy in the prediction of the confirmed cases and fatality.