I am working on trying to make an accurate model for predicting COVID-19 growth based on the per-county figures we have with COVID-19 (courtesy The New York Times). To say the data is noisy would be an understatement.
What I have found, so far, is that if we look at current daily growth (averaged over seven days) and use exponentiation to predict future growth based on the previous week’s figures, the numbers are too high (usually by a factor of two, but the error amount is all over the place).
Point being, we’re seeing a more complicated growth model than simple exponential growth; the actual growth is lower.
This is a work in progress and I’m nowhere near being able to make a simple easy to read graph showing a reasonable projection of COVID-19 growth in the United States.
Basically, it uses the fact that mean time from contagion to death is 17,3 days. For France, it gives reasonably accurate predictions. Best than almost any other model I've seen, in fact.
It could only be exponential if the population was infinite. It has to be more of a bell curve, because as infection grows, there are fewer hosts to infect, and therefore growth would start to decline.
Even in places where 0.05% of the population has confirmed cases of the virus, I’m seeing the curve flatten. Whether that’s from quarantine of from the virus hitting its limit, I can not say. But the curve is flattening.
Yes that's good news. However when you say the curve is flattening, I believe you mean we have reached peak daily new cases, meaning the number of new cases we see is going to start declining from here.
Collectively many people seem to be referring to that as "curve flattening", but my understanding is that flattening the curve means slowing the growth rate overall, so that it takes us longer to reach peak daily new cases. It is not intended to indicate a particular point along the x axis. In fact if we are actually flattening the curve, it will take us LONGER to reach our peak. Also, its difficult to measure whether we have been successful or not, because the only thing we have to measure against would be hypothetical worse case scenarios.
What I have found, so far, is that if we look at current daily growth (averaged over seven days) and use exponentiation to predict future growth based on the previous week’s figures, the numbers are too high (usually by a factor of two, but the error amount is all over the place).
Point being, we’re seeing a more complicated growth model than simple exponential growth; the actual growth is lower.
My work so far is on GitHub: https://github.com/samboy/covid-19-html
This is a work in progress and I’m nowhere near being able to make a simple easy to read graph showing a reasonable projection of COVID-19 growth in the United States.