1/ looking at covid stats is tricky bc - noisy day-to-day data - weekends are different from weekdays, for whatever reason I have been tracking US covid deaths per day, and smoothing with both a 3-day running average and a 6-day running average
-
-
3/ The good news is that the VAST MAJORITY of days display that characteristic. The bad news is that as times goes on, more and more days fail that characteristic. If deaths is the data, growth rate is the first derivative, and growth of growth rate is the second derivative.pic.twitter.com/VBjaGjP5LE
Show this thread -
4/ Second derivative is still good ... but third derivative is not so good. Will be interesting to see where this goes. My prediction is that within 2 weeks we're going to see more days than not fail this test.
Show this thread -
5/ To be clear, all of this is not a "model". Models aren't just curve fitting and tweaking params and saying "Eureka! I understand it". This is just playing with data, trying to get a handle on what the derivatives look like. Also: my thoughts are worth what you pay for them
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.