10% of men without and 30% of men with test positive for a certain disease. What are the false positive and false negative rates of the test?
Out of all men... \(10\%+30\%=40\%\) test positive, be it false or true.
Thus \(100\%-40\%=60\%\) test negative, be it false or true. We want to know which of these are false though.
I'm not sure if 10% would be the false positive rate. It states that 10% of men without the disease test positive for it, not that 10% of men who take the test are falsely diagnosed
Is this a probability class?
yes
Okay, let's say \(A\) is the event you have it. \(B\) is the even you test positive.
\[ \Pr(A^C|B) \]This represents false positives, right?
yes, that makes sense to me
\[ \Pr(A^C|B)=\frac{\Pr(A^C\cap B)}{\Pr(B)} \]
Now you claimed that \(10\%\) would be \(\Pr(A^C\cap B)\).
I believe that \[ \Pr(B) = \Pr(A^C\cap B)+\Pr(A\cap B)=0.10+0.30=0.40=40\% \]
So \[ \Pr(A^C|B) = \frac{0.1}{0.4}=\frac 14=0.25 = 25\% \]Does this make sense. Can you buy that?
Doesn't make a whole lot of sense somehow.
I think \(\Pr(A^C\cap B)=10\%\) is rate of false positives. Remember that \(10\%\) is of the total population, not only of those who are tests. You tricked me man.
Yes, it's not 10% of the men tested. But it isn't 10% of the total population either, it is 10% of the population that doesn't have the disease.
So \(\Pr(B|A^C)=10\%\)?
Thats how I'm understanding the questions
\[ \Pr(B|A) = 30\% \]
I almost feel like there isn't enough information. I think that if we knew the probability that a man would have the disease then it would be much easier
|dw:1381651233560:dw|
Join our real-time social learning platform and learn together with your friends!