Estimating the Final Epidemic Size for COVID-19 Outbreak using Improved Epidemiological Models

Rajesh Ranjan

In this work, we estimate the final epidemic size for some countries which has been hit badly due to this outbreak. Since, the results from epidemiological models can be dependent on its implementation to estimate parameters, initial condition, etc. even for same underlying data, we use three different models to compare the outcome. Apart from the logistic and simple SIR model, a generalized SEIR model is used that takes into account of quarantined and deceased patients. Further, we introduce a correction to the estimates from these models in order to account from slow decay of infections as seen in Fig. 3 for the US and other European countries. Section 2.1 presents a brief discussion of these models: details of their implementation can be found in the corresponding references given. The corrections introduced to account for slow decay of infections compared to those observed in China are discussed in section 2.4. Section 3 shows the predictions for some key countries and a brief discussion is included. It is not possible to include the estimates for all the countries in this paper. However, it is expected that the reader can make use of these models, codes for which are available in the open-source, to estimate the final epidemic size for regions not included in this paper. As a caution, it should be mentioned that if the reported number of cases begins to exceed the predicted end-state systematically, then the epidemic will enter a new stage, and a new estimate needs to be made incorporating the effects of the changed parameters.