[conspire] COVID-19: wee bit 'o math, and ...

[Rick Moen]

Quoting Michael Paoli (Michael.Paoli at cal.berkeley.edu):

> I think some of the better bits covering the math, and "exponential"
> (actually logistic - at least in theory) curve, etc.

It becomes a logistic function ('S' shape or sigmoid curve) eventually 
after starting to kill lots of people and saturating the population.

The early to medium part of that logistic curve, up through the middle
of where mass deaths occur but before the population approaches
saturation, is exponential.  

The reason people speak of the problem being exponential growth, rather
than being achingly precise and saying 'actually logistic - at least in
theory', is that we're attempting to flatten it out long before the mass
deaths and eventual leveling off as described by mathematician Pierre
François Verhulst in a series of three papers between 1838 and 1847 as
the _eventual_ end-result of exponential population growth limited by
eventually finite limiting resources.

Seriously, Michael?  You really think fine points about extrapolated
growth graphs are a good contribution to this thread?  Well, better you
than me, sir.  {grumble}


> but if we presume to a (rough) first degree approximation that that
> was and is the case, then I'd expect to see decrease in rate for the 6
> counties, right around 5 days after the relevant orders went into
> effect ... 

There are a few jokers in that pack.  One, without an adequate
supply of testing kits, and deployment rules at hospitals and counties
that stop hoarding them, the statistics are not accurate in the first
place.  Two, you can only very roughly guesstimate the real case counts
from the confirmed case counts.  Three, unless you have good data about
the degree of compliance in the field to the state and county protective
orders, you are going to have a very difficult degree predicting in real
time how much the growth rate has been slowed.

If those jokers have been worked around in some reliable way, I haven't
yet heard how.

Anyway, I respect your trying to do prognosis based on available
numbers, but urge bearing in mind the difficult-to-control factors that
make that difficult.



> Treatments?  I dunno, haven't been following that closely - lots of bits and
> pieces of "reports" out there of varying credibility.

'Possible new COVID-19 treatment found by blah-blah researchers!' is a
winning news-clickbait formula, ergo we predictably get a lot of it, and
then we get to waste time trying to get the real story.  Personally, I
would rather disregard all of that as an obvious time-sink.  When
something gets meaningfully validated instead of just being a breathless
and paper-thin bit of pseudonews, we'll hear about that.

-- 
Cheers,                                "Rand Paul being patient zero for a Senate 
Rick Moen                               viral outbreak is a sign of a writers' room 
rick at linuxmafia.com                     dropping too much acid, late in the season."
McQ! (4x80)                                           -- @owillis (Oliver Willis)