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

Rajesh Ranjan

Typically, epidemiological models such as a logistic or SIR model is preferred for modeling the later stages. However, these models are highly dependent on initial conditions and underlying parameters, and an incorrect estimation of these can give completely different results. Further, modeling these only on data from an exponential curve can be misleading in estimating the final epidemic size. However, when there are signs of decline in the exponential growth (t 1 in Fig. 3 ) and sufficiently long data is available, these models can be reliable for predictions within uncertainty limits. Though, waiting for the curves to show the sign of flattening, can be relatively late for making policy decisions. Nonetheless, a reliable estimate of the final epidemic size based on these data can be very helpful for taking necessary measures in terms of building adequate healthcare infrastructure and early preparation for the future. Further, it is known that 4 . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.