Re: We're all doomed! Doomed I say- the Corona virus thread for panicking in!

[azriel]

https://covid19.healthdata.org/projections
(Amarië was here.)

This didn't show up for me ? Sorry Halfy.

[halfwise]

It seems to have picked up some extra characters. You can go back and click on my post which I edited, or just click on your again and edit out anything that obviously doesn't belong up there in the browser bar.

[halfwise]

So I'm curious Halfy, what are the assumptions on which the decline following April 15 are based?
It would seem to me there are so many unknowns at this point, and so many variables that we're right now in the process of tweaking, that it would be hard to make a meaningful prediction at this point.
So I'm assuming some fairly broad assumptions. No?

It's actually pretty simple.  In my model I only had three things going on:

1. the infection rate per relevant interaction was assumed constant. I got this from the initial exponential growth before #2 and #3 comes into play.
2. the assumed rate of relevant interactions is proportional to the density of people infected multiplied by the density of people who had never been infected, or  (current infectees)x(Population - accumulated infectees).   It's this interaction rate that is variable due to both changes in the acquired immunity, and differences in human behaviour.  This human behavior constant is rolled up with the infection rate per interaction as a single proportionality factor.
3. People recover 14 days after acquiring the disease, and are no longer infectious plus have immunity.

So the decline comes from the population being saturated (everyone is assumed to become infected) plus the recovery time.  My simple model assumes everyone behaves the same, so everyone will eventually get infected.  More sophisticated models will have a range of behaviors, with some people staying uninfected throughout the local explosion, but will be susceptible to later infection from outside due to travel from localities that are late starters.  This will stretch out the tail of the recovery.  My simple model is symmetrical but still catches all the main features but that one.

As far as assumptions go, the big one is that relevant human interactions go as the product of infected and uninfected densities.  The proportionality factor will change: I saw the slope on a semilog plot decrease by a factor of 2.5 after New York declared social isolation.  It then remained remarkably constant (the constancy in slope before and after the decrees made my jaw drop), but I would expect a progressive change as the number of cases becomes noticeable on a more human level: seeing that neighbors and friends are getting infected.  You don't notice when less than 1/1000 are infected, but do when it reaches 1/20.   This would have the effect of flattening the curve more: stretching it out and lowering it.  The question is, how much?  I suspect this second flattening is not taken into account in the models because it's totally unknown PLUS their timing agrees with mine, and I haven't put that in there.  It could be a large effect: at the end of this week I plan to make my last trip to the store for several weeks, and I expect I'm not alone.  I could assume an adjustment, but it would be complete guesswork.  I've seen the Italian rate flatten out, but I don't know if it's real or due to sick people not bothering to get tested because what's the point with an overwhelmed health care system?

[azriel]

Thanks Halfy

[David H]

[quote="halfwise"]

It's actually pretty simple. In my model I only had three things going on:

1. the infection rate per relevant interaction was assumed constant. I got this from the initial exponential growth before #2 and #3 comes into play.
2. the assumed rate of relevant interactions is proportional to the density of people infected multiplied by the density of people who had never been infected, or (current infectees)x(Population - accumulated infectees). It's this interaction rate that is variable due to both changes in the acquired immunity, and differences in human behaviour. This human behavior constant is rolled up with the infection rate per interaction as a single proportionality factor.
3. People recover 14 days after acquiring the disease, and are no longer infectious plus have immunity.

That all makes sense in a city. You get to assume that everything is uniform. We could do the same for our county with a population of 20,000, with different numbers for the ways we interact. The problem is that we've only had 27 people tested and no positives so far, yet we're only a couple hours drive of Seattle/Tacoma where they've had 25,000 tested and 3500 positives.

For us, that curve in the city looks like a giant breaking wave that's hanging over our head, ready to crash down on us and overwhelm us. How would you model that I wonder.....?

[Mrs Figg]

https://covid19.healthdata.org/projections
(Amarië fixed it. Amarië is cool.)

This didn't show up for me ? Sorry Halfy.

yeah me neither


also a doctor in Korea has said that it has been found that some people can get it twice, so are not becoming immune once they have had it.

[Lancebloke]

Well.... its been a while but I have the whole evening. And the next. And the next....

[halfwise]

Did you try again after Amarie fixed it?  But it doesn't do Europe anyway, sorry.

I hope the news from North Korea is wrong.

https://covid19.healthdata.org/projections
(Amarië fixed it. Amarië is cool.)

This didn't show up for me ? Sorry Halfy.

yeah me neither


also a doctor in Korea has said that it has been found that some people can get it twice, so are not becoming immune once they have had it.

[halfwise]

I keep hoping that youtube will bypass artificial national barriers. Not sure this will get through.



It's all about the italian politicians yelling at their citizens. go to 1.45.

[Mrs Figg]

That doesn't play for me, but I think it is this one I posted earlier, Italian mayors going ballistic at seeing idiots in the streets.

[azriel]

Sorry Halfy, its not playing for me, which, considering how serious this is, youde think it would ?
Gotta love the Italians ! They say it like it is

[halfwise]

It had some of those, a few others.

Interesting - here we are allowed to go out so long as we stay 6 feet away from people, but in Italy it seems to be the rule not to go out except for dire need.

[halfwise]

Starts out a bit slow, but turns into high comedy:

https://www.marketwatch.com/story/astrophysicist-hospitalized-after-getting-four-magnets-stuck-in-his-nose-while-making-a-coronavirus-device-2020-03-30

[Mrs Figg]

It had some of those, a few others.

Interesting - here we are allowed to go out so long as we stay 6 feet away from people, but in Italy it seems to be the rule not to go out except for dire need.

yep we are not allowed out unless to buy food or emergencies. Yesterday we had a lorry with a loudspeaker driving up and down the mountain telling everyone to stay at home. They are taking it very seriously even in hamlets like mine.

[Pettytyrant101]

{{ One of the interesting aspects to come of all this is that we in the UK at least, and seems similar elsewhere, are in a state of martial law, a police state effectively where you can be stopped and questioned just for leaving your own home.
A state of affairs I thought Id never see, let alone one in which the people are not only willing to go along with it, but actively want it.
Fear is a strong motivator for a large group to hand all the power to a small controllin group.

What will be interesting is to see how much the government hold onto when its all over. After all there are several laws on the UK books that have been in common use for decades, but have their source solely in being emergency powers invoked during WW2 the government then never unrevoked. }}

[halfwise]

I even know a die-hard libertarian who for once is saying the government should do more. I've found that people are reacting with surprising rationality, given what dumb-shits Americans often seem to be. It's nice to know that when it really matters people tend to recognize and accept the right thing to do.

[Pettytyrant101]

{{ Dont worry Halfy, youre still hanging onto your Americanisms in some aspects- I see some states are insisting that gun shops are vital services that have to stay open! }}

[halfwise]

Sure - given the choice of slow suffocation without a respirator and a quick shot in the head, which would you go for?

I hear some folks in Pennsylvania are experimenting with funnels taped to leaf blowers. While high paid doctors and captains of industry a fussing around with how to supply ventilators, country folks get the job done.

[Amarië]

Da fuk?! That is some dark and sick Old Yeller logic.

Free healthcare. That's what I'd go for. A government trying to make sure all who needs respirators will get one. Not just the ones who can pay. And having heard about the clean up after a gun to the head suicide... Nope.

[halfwise]

But the clever thing is, the suicide never has to clean up the mess. Whether that makes it clever or a cop-out I shall forbear to judge.

[azriel]

[azriel]

[David H]

{{I see some states are insisting that gun shops are vital services that have to stay open! }}

Of course! Because how are gangs of armed vigilantes going to police the highways streets and roads after the zombie apocalypse if they can't buy ammo?

{{I'm sure some of my neighbors are already well stocked, just in case.... }}

With respect to Halfy's observation, I think that on our side of the continent the preference would be for booze and pills over both those other alternatives. Probably regional cultural differences...

[azriel]

https://www.facebook.com/BBCWalesNews/videos/2604920209760848/

[halfwise]

Have to say I'm proud of how the New York citizenry is handling things. The grocery store right across from me started to accommodate a few days ago first by marking spaces between those waiting to check out with tape on the floor, then they put up plastic screens to protect the checkout clerks, and now they are limiting the number of people inside to keep the density down. This is a neighborhood of immigrants, largely hovering in the area of poor. But they understand the protections and quietly adhere to the guidelines. Just walked by a line of them waiting to be let in the store, spaced out and accepting the situation.

Everyone has seen the news of tents being set up in Central Park for overflow from Mt Sinai hospital, and the Navy Hospital ship that arrived yesterday. People understand the consequences and are playing by the rules.