23 Mar 20
@eladar saidThat was Ponderable, not Deepthought
@DeepThought
I am willing to bet that it will be more accurate than your exponential growth equation after 1000 days.
@deepthought saidI went ahead and did the full data set for US Deaths: ( I realized it was flawed as the origin of US Deaths was not until March 1. With the exceptions of ( 0,0 ) and ( 1,1 ). If I run the regression with those points I get a Singular Matrix. So I just took out rows until I got a result.
A couple of observations, let's make the independent variable t for time. Let c --> c' = -c, and b = exp(ct₀), we now have:
L[t] = a/(1 + exp(-c(t - t₀))) + d.
Let's work this out for some values of t.
L[-∞] = d
L[0] = a/(1 + exp(ct₀)) + d.
L[t₀] = a/2 + d
L[∞] = a + d
Infinitely far in the past we expect the virus not to exist. In other words we expect d ...[text shortened]... Your estimate for a and d gives total deaths of 2,362 which is looking very optimistic at this time.
Data: ( same source as before: worldometer)
2 6
3 9
4 11
5 12
6 15
7 19
8 22
9 26
10 30
11 38
12 41
13 49
14 57
15 68
16 86
17 109
18 150
19 207
20 256
21 302
22 419
a = 4501.485
b = 4697.06664
c = -0.278611475
d = 10.519
The IP March 31st, 2300 Deaths.
Max hitting 4500 Deaths by Day 49 ( April 18th )
@eladar saidWell, eventually you'll get more deaths than people, the logistic function has the benefit of saturating.
@joe-shmo
I am thinking in a month is enough time to feel confident.
The only thing I am totally confident about is that an indefinte exponential model will overestimate the number of deaths.
23 Mar 20
@DeepThought
Since this virus does not even kill 3 out of 100 actually infected, I think it is safe to say the number killed by this virus will be well under the world's population.
24 Mar 20
@eladar saidIt may be less than that. The only way to know is to test everyone and we all know that is impossible. Unless you know how many people are infected you cannot possibly know the correct percentage of the death rate. Some people have mild symptoms and may dismiss it as a cold. They were never tested and are left out of the statistics.
@DeepThought
Since this virus does not even kill 3 out of 100 actually infected, I think it is safe to say the number killed by this virus will be well under the world's population.
@metal-brain saidIf you wanted to do it, you could take a random sample of people and see if they are sick or have the antibodies. Then you can get an estimation of percent who have the virus.
It may be less than that. The only way to know is to test everyone and we all know that is impossible. Unless you know how many people are infected you cannot possibly know the correct percentage of the death rate. Some people have mild symptoms and may dismiss it as a cold. They were never tested and are left out of the statistics.
But they have other things to worry abput at the moment.
@Metal-Brain
So far in the US it's clocking in at about 1.5 % fatalities. That is still ten times higher than the last epidemic.
@ponderable saidFor clarfifcation the curve was calculated for the cumulated number.
Taking the data from
https://www.worldometers.info/coronavirus/country/us/
which report 256 dead as of yesterday (and who give sources) and their daily number starting with one dead at 29th of February I can model the curve death over day using the exponential curve
1.61*exp(0.24*d) with a r-value of 0.97 which is reasonable.
If I calculate to day 42 (which is three we ...[text shortened]... Worldometers.info just changed to 276 death today it should be around 316 if the curve was correct.
And I put in the last three data points and the r-value actually went up to 0.99, so we do observe (at least for the time being) an exponential growth, and I really expect this to go on for some time.
@Eladar
The UK the number of deaths has been approx 1.4^(number of days) over the last 2 weeks.