Glanced through it so apols to all concerned for any misintetptetation on my part. Many points there so will comment on just 2.
1) Gompertz: Agree & disagree. The death curves for most locations are gompertz-like but not 'perfect' Gompertz, due, for example to effects of 1/n
2/n. time lag to death, which kinks that now-famous near-linear log(daily/cumulative) deaths plot. Happy to discuss that with any of the technical guys. I think the key point with Gompertz though is that what we have here appears to act as a natural process - which it does.
...
3/n 2) Sweden: here's some work I did looking at the UK & Sweden. Please understand it though. The fact that Sweden and UK both have near identical deaths isn't as interesting as the fact that they don't have very different deaths.
...
4/n Because, if the "high fatality & everyone is susceptible" hypotheses is correct then outcomes would be highly sensitive to the extract situation & to, e.g., timing.
Illustrative example:
...
5/n. So even one believed that Sweden did effectively lock down to some degree, their outcome would still be very unlikely to be so similar to UK & the other high fatality locations.
The similarity in outcomes between the high fatality locations is signifiant.
...end.
PS Please forgive typos. Mobile phones....
NB. This thread was written after reading this tweet:
NB. This thread was written after reading this tweet:
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.