OK so thre's a disease that kills 1 in a million. They develop a test that give you a 99% accurate reading either way...

OK so thre's a disease that kills 1 in a million. They develop a test that give you a 99% accurate reading either way. 1% of the test is wrong.

You get your results back and it comes back positive. What are your chances of having said disease?

Useless test. More people will be diagnosed positive than people that actually have it...

I wanna say 99% but I'm probably a faggot

1 in a million, you're not clever

The test has come back positive though

Are you a statitician?

And it's a 99% chance you have it

indeterminate, how man people who get the disease die from it?

Always fatal when diagnosed correctly. 100% mortality rate.

No but 1% of 1 million tested people is 10 thousant positive resuts. But only 1 of them is infected (statistically speaking)

OP here. You get the disease you will die a slow and horrid death 100% if you contract it. No cure available

99/1000000

Winrar. Even with a positive test that's 99% accurate you're more likely to be clear of the disease

I'd still shit myself though

The probability of actually having said disease is 99%. You already laid out that the test is 99% accurate.

It's irrelevant how many people die from it, and how often it is fatal.

Plus, what this guy said:
It's a horrible test.

The actual mortality rate is irrelevant to the question. If the test has a 99% accuracy rate, the chances that you have the disease when said test comes back positive (the question asked) is 99%.

99/999999 dumbass

no its a 1 in 10,000 chance even with a positive test

wait, it's 99/1000098

>Sup Forums - Random
>The stories and information posted here are artistic works of fiction and falsehood.
>Only a fool would take anything posted here as fact.
>[Post a Reply]
>10/04/16 New board for Sup Forums Pass users: /vip/ - Very Important Posts
>06/20/16 New Sup Forums Banner Contest with a ch
Where did u get that figure

well, 99%? It says "chances of having" not "chances of dying from it"

Do the test twice.

Interesting. I wonder how this would skew the stats

It is true that for each 1M tested 10000 will be false positives, but if the test is 99% accurate and it comes back positive you only have 1% chance of being one of those 10000. The chances of having the disease are 99%

If you get it you shit out your organs. Fatality rate is 100%

That makes no sense nigga

Even with a 99% accurate test it will still make 10000 false positives. Are you more likely to be the 1 or one of the 10000

If 1M people that are infected take the test, only 10000 will come out negative, on the other hand if 1M people that are not infected also take the test only 10000 will be false positives.

The test has 99% accuracy, doesnt matter how many people are tested or how many die, if the results are positive you hate a 99% chance of being infected.

Assuming the likelihood of having the disease is 1:1 Million
P(sick)=1/(10^6)
P(test=pos|sick) = 99/100
P(test=neg|sick) = P(test=pos|not sick) = 1/100
P(not sick) = 1 - P(sick)

Total likelihood: P(test=pos) = P(test=pos|sick)*P(sick) + P(test=pos|not sick)*P(not sick)

Bayes Formula: P(sick|test=pos) = P(test=pos|sick)*P(sick)/(P(test=pos))

It's mathematically certain that you more than likely do not have the disease. Even with a positive test

So if you get the possitive result what is the answer to the equation?

So every person has a chance of "1 in a milion" of having the mentioned disease. Now after taking the test and coming out positive, you have chance of 1% of "1 in a milion"=> 1 in 1000 is that you have it

1 in 10000

50000% chance because OP is a fag

yeah, i fucked up, thanks for correcting

if my calculator knows shit then its approx 98,9903.. * 10^(-6)
So it's very unlinkely
You can easily calculate it on paper. It's just addition and multiplication

Get a positive result
Numbers say not to worry
Die anyway from stress
100% chance of death

the tests that didnt work didnt work for a reason.
prohably has to do with some qualities of that person so you would maybe get the same results everytime (or at least with different prohabilites)
this simple statistical model is just not good enough to make any kind of qualified estimation for this