SIR Model Adjustment for Covid Spread in China

Abstract

One of the ways to reduce the harm caused by COVID-19 pandemic is developing policies that balance the health and economy of society. These policies rely on large amount epidemic data. Traditional epidemiological SIR models become the basis for COVID-19 models to replicate and predict the epidemic's trend. This study seeks to find out which model is more suitable for countries under strict control policies. Through the collection and analysis of information on the epidemic in China, this study concluded that there are two pieces of data are suitable for this study. In this study, the two parts of data are calculated and combined with other studies to obtain the parameter values ​​for different models. The SIR and SEIR models in this study yielded interesting results. Simulations of the SIR model under mass lockdown policies produced reliable data forecasts for infection days. The SEIR model has a relatively accurate prediction of the trend of the proportion of infections population. However, other models based on COVID-19 characteristics do not produce as much information as compared to the above two models.