The attached notebook takes fatality resource data and creates a curve fit based on the data and the fact that the expected fatalities will reach constant in the large time limit.
When assuming that the reproduction rate R is not a constant, that we are still in the beginning of the pandemic (i.e. far away from herd immunity), then R(t) can be estimated.
The model shows that social distancing has worked lowering R from 3 to about 1. It also shows that the Expected number of fatalities (USA) by the end of June will be over 200,000. Significantly higher than the current IHME models predict.
Thank you for the badge!
I posted this because I think the IHME Model is way off and overly optimistic. They constantly have to upward revise the fatalities.
This post has been listed in the main resource-hub COVID-19 thread: https://wolfr.am/coronavirus in the section Computational Publications. Please feel free to add your own comment on that discussion pointing to this post ( A simple estimate of covid-19 fatalities based on past data ) so many more interested readers will become aware of your excellent work. Thank you for your contribution!
– you have earned Featured Contributor Badge 
