Post
Topic
Board Speculation
Merits 1 from 1 user
Re: Wall Observer BTC/USD - Bitcoin price movement tracking & discussion
by
fluidjax
on 15/02/2020, 17:57:57 UTC
⭐ Merited by vapourminer (1)
Oh and in regards to lethality: Calculate deaths/(deaths+cured).

Don't know why people keep using total number of infected. It makes no sense. Some of those people are going to die.

Because death occurs earlier in the disease cycle than cured state, (cured may be pronounced many weeks after the onset whereas death in a couple of weeks)
and especially at the beginning of the outbreak this figure will be weighted too far towards a high mortality rate, and equally makes no sense at this point.

You can easily see the error of using this method in the first few weeks, say 5 people died, but nobody has been declared cured, giving a 100% mortality rate.


Bear attempts look pretty futile, nobody else appears to be jumping aboard.
Again, some of the infected are going to die. And we don't know how many of them. We do know how many died and how many were cured. Convenience is no excuse for deliberately fudging the numbers. And it's not 5 people anymore.

I don't dispute that the numbers are fudged, I'm disputing that your alternative method to calculate the mortality rate is also wrong. Even if the available figures were correct, which they are not, there is no way to calculate an accurate mortality rate.Most likely, it lies somewhere between your value and the media's  value, with such a large range it's not very helpful.

(5 was a made up figure, to illustrate the error in the methodology.)

I don't see the problem. Infected people will either end up in the dead category or the cured category. What's missing?

People die quicker than they are declared cured. This difference means that at any point in time the dead is known, but the number who will be cured lags by weeks, weighting the statistic towards a higher mortality rate.