Citat:
Mycket bra förklarat!
Från en intelligent postare på Reddit:
"I posted a similar thing yesterday and wanted to add some information on the mortality rate and binomial distribution.
Binomial distribution calculates the probability given the chance and number of successes. Example: Rolling a standard 6 sided die (plural is dice) has the probability of picking a number (say, 4) and rolling it of 16.7% (or 1/6). Picking that number, rolling the die twice and having that number come up both times is a 2.8% chance, or having it come up once on either roll a 27.8% chance.
So far nobody has died outside Mainland China with 153 cases. Given the 2% mortality rate (think of it like a 50-sided die with 1 marked death and 49 life that is rolled once for each case), the chance that none of the 153 cases outside Mainland China would result in death is 4.5%. Certainly a possibility, but not likely.
On a similar note, 10 people outside of Hubei province have died out of 4,221 cases not in Hubei province. The chances (cumulative probability) of 10 or fewer people dying with a 2% mortality rate and 4,221 cases is extremely small: 6.40E-23 % or in odds terms less than one in a quintillion chance of happening.
A couple other ways to expand on it: Checking the chance of 20 or fewer deaths with a 2% mortality rate is still 2.38E-15 or about one in a quadrillion chance of happening. Going by the number of cases outside Hubei province from 3 days ago (2,493) to account for time for the virus to cause death , the odds of having 10 or fewer deaths with a 2% mortality is less than 1 in 100 billion, the odds of it being 20 or fewer about 1 in a million.
On the flipside, 10 deaths out of 4,221 would give a fatality rate of 0.23% and 10 out of 2493 about 0.4%.
Long story short, I think the actual mortality rate when an area is not under a large outbreak that strains the local health centers (and also now that we have a better idea of what supplies are required when there is an outbreak) is going to be well under 2%, I'd even venture it to be well under 1%."
https://www.reddit.com/r/news/comments/evvzf9/coronavirus_megathread/fg64kdl/
"I posted a similar thing yesterday and wanted to add some information on the mortality rate and binomial distribution.
Binomial distribution calculates the probability given the chance and number of successes. Example: Rolling a standard 6 sided die (plural is dice) has the probability of picking a number (say, 4) and rolling it of 16.7% (or 1/6). Picking that number, rolling the die twice and having that number come up both times is a 2.8% chance, or having it come up once on either roll a 27.8% chance.
So far nobody has died outside Mainland China with 153 cases. Given the 2% mortality rate (think of it like a 50-sided die with 1 marked death and 49 life that is rolled once for each case), the chance that none of the 153 cases outside Mainland China would result in death is 4.5%. Certainly a possibility, but not likely.
On a similar note, 10 people outside of Hubei province have died out of 4,221 cases not in Hubei province. The chances (cumulative probability) of 10 or fewer people dying with a 2% mortality rate and 4,221 cases is extremely small: 6.40E-23 % or in odds terms less than one in a quintillion chance of happening.
A couple other ways to expand on it: Checking the chance of 20 or fewer deaths with a 2% mortality rate is still 2.38E-15 or about one in a quadrillion chance of happening. Going by the number of cases outside Hubei province from 3 days ago (2,493) to account for time for the virus to cause death , the odds of having 10 or fewer deaths with a 2% mortality is less than 1 in 100 billion, the odds of it being 20 or fewer about 1 in a million.
On the flipside, 10 deaths out of 4,221 would give a fatality rate of 0.23% and 10 out of 2493 about 0.4%.
Long story short, I think the actual mortality rate when an area is not under a large outbreak that strains the local health centers (and also now that we have a better idea of what supplies are required when there is an outbreak) is going to be well under 2%, I'd even venture it to be well under 1%."
https://www.reddit.com/r/news/comments/evvzf9/coronavirus_megathread/fg64kdl/