John Phillips
pfm Member
You haven't published the basis for understanding what you have fitted and how. Below is an alternative view with all of the details to reproduce the analysis and see if you agree or not:I'm get about a 14 day doubling time on admissions matching the rise in cases. Vaccines haven't yet broken the link as far as I can see...
You can get 14 days if you pick the most recent data and take the first fitted point from the very low and the second point from the very high. On a longer view as above there seems to not be an indication of statistically significant recent acceleration of the 29 days best fit to 14 days. The 29 days is loge(2)/(the exponential 0.0242 of the fitted curve). The data on vaccine effectiveness against Delta suggests we should not be seeing a significant acceleration in admissions as they increasingly come from Delta.
However I am DEFINITELY NOT saying your 14 days is wrong. But other answers are possible from the data we have.
Would you reveal what data set you used (UK or England, for example), what day was day 0, and how many points you fitted to your exponential curve.
The analysis of noisy data is uncertain and it is IMHO unwise to ignore the uncertainty in either optimistic or pessimistic analysis and insist that one answer is definitive. It is essential to deal with the likely range of possibilities.