my roommate was a math major and took that class, they did a bunch of obscure abstract proofs like "prove numbers exist" or some stupid shit, the kid would do his homework and have 5-6 pages for a single proof.
I'm in "that class" right now...
my roommate was a math major and took that class, they did a bunch of obscure abstract proofs like "prove numbers exist" or some stupid shit, the kid would do his homework and have 5-6 pages for a single proof.
These FP and FN are both WRONG answers, but depending on the relative number of FP to FN, the answer to raven's question is different. The parameter raven asked for is known as the true positive rate, TPR, which is defined as the number of true positives out of the total number of positives identified by the test.
Mathematically that can be written as:
TPR = FP / (TP + FP)
However, FP is not uniquely defined for a given value of acc. A second parameter that is independent of acc must be defined in order to determine FP and hence TPR. Based on the value of FP, TPR can range from 0 to 1, the full domain for TPR.
And I wasn't even a math major.
this isn't true - FP is something you can measure. for instance, in the real world, you can do mammography that gives a positive test for breast cancer. then you do a biopsy and realize it's not breast cancer. that is a false positive.oh, i get it; and like i said before, unless its explicitly stated, i've always treated the accuracy such that the FP is essentially 0. So for the purposes of raven's posed question, i have been treating the "false correct answer" probability as 0.
in the real world, if you have a device that is 95% accurate, theres no way to measure its FP. this device gives the right answer 95% of the time and the wrong answer 5% of the time. maybe its only 90% accurate and 5% of the time it is giving the right answer even though its a 'false right answer'. but theres no way to determine that. so if you can guarantee it gives the right answer 95% of the time, regardless of how much of that percentage is a 'FP', its 95% accurate.
What you are saying is correct and totally makes sense, and you can nitpick on the theoretical side all you want, but this sounds like an argument you would only deal with in a textbook.
Why the hell is this thread 29 pages? Did Kahula show up in here or something?
this isn't true - FP is something you can measure. for instance, in the real world, you can do mammography that gives a positive test for breast cancer. then you do a biopsy and realize it's not breast cancer. that is a false positive.
in general, you need a "gold standard" test that is 100% accurate to know the TRUE state. you don't use it because it's expensive, inconvenient or whatever so you have cheaper, more convenient tests with less accuracy that you use for screening.
if FP=0, then the answer to raven's question would be 100%.
Code:| Mormon | Not Mormon ------------------------------------------------------ Test Positive | [b]True Positive[/b] | [b]False Positive[/b] ------------------------------------------------------ Test Negative | [b]False Negative[/b] | [b]True Negative[/b]
Total False = FP + FN = 5% * population
a) this means you can't know FP without knowing more information.
b) FP and FN are interrelated by the equation above, so you cannot just look at FP.
depending on the ratio of FP/FN, the answer to the question could range from 0% to 100%.
this isn't true - FP is something you can measure. for instance, in the real world, you can do mammography that gives a positive test for breast cancer. then you do a biopsy and realize it's not breast cancer. that is a false positive.
in general, you need a "gold standard" test that is 100% accurate to know the TRUE state. you don't use it because it's expensive, inconvenient or whatever so you have cheaper, more convenient tests with less accuracy that you use for screening.
if FP=0, then the answer to raven's question would be 100%.
I'm in "that class" right now...
If FP = 0, then the overall accuracy of the test could not possibly be 95%. A test that returned negative on every single person would have FP = 0 and an overall accuracy of 98.1% since only 1.9% of Americans are mormon. Are you really this fucking dumb?
I'm wondering if you were really too dumb to interpret the question or if you are just a bad troll
what is going on in here?
a) should be assumed if only the accuracy is given flatly as "detects accurately 95% of the time"; therefore 5% rate for FP, and 5% rate for FN;
b) yea like i said id always assume they were the same unless explicitly stated in the problem.
They hijacked my thread about 10 pages ago to turn this into statistics discussion. What we are finding out is that there are a lot of people who can't do statistics worth a crap.
There is absolutely no reason to assume the test is equally good at detecting Mormons as it is detecting not Mormons. In general, it is NOT true for a test. Moreover, we cannot even consider it as a conservative estimate since without making the assumption the answer can take on all possible values in the domain.
This isn't anything like an approximation. An approximation would be saying, well a supercomputer can do this numerically and come up with 0.0009, but we can do it with simple algebra and get 0.001. This is saying the answer is ANYTHING, but for argument's sake we'll say it 25%. You don't seem to realize that this assumption is not reasonable AT ALL.lol are you daft?
he made the problem up, theres no such thing as a mormon detector. its a hypothetical.
And there is absolutely no reason to assume it isnt equally good at detecting FP and FN seeing as he didnt explicitly state otherwise. i am glad you found something to argue about with ravens problem which defends you telling him he is incorrect, but now this is getting stupid.
Do you/did you sit there in class and nitpick with a professor when they made an approximation? Do you/did you sit there during a test and not answer a test question because every assumption wasnt explicitly written out? jesus, you wouldnt survive a day in an engineering class.
ps: i am done here. SoD you should update this thread after your second date when you start bleeding from the mouth and she tells you to have a nice life.
lol are you daft?
he made the problem up, theres no such thing as a mormon detector. its a hypothetical.
And there is absolutely no reason to assume it isnt equally good at detecting FP and FN seeing as he didnt explicitly state otherwise. i am glad you found something to argue about with ravens problem which defends you telling him he is incorrect, but now this is getting stupid.
Do you/did you sit there in class and nitpick with a professor when they made an approximation? Do you/did you sit there during a test and not answer a test question because every assumption wasnt explicitly written out? jesus, you wouldnt survive a day in an engineering class.
ps: i am done here. SoD you should update this thread after your second date when you start bleeding from the mouth and she tells you to have a nice life.
pps: if the last 6 pages were deleted, this thread would still be humorous.
This isn't anything like an approximation. An approximation would be saying, well a supercomputer can do this numerically and come up with 0.0009, but we can do it with simple algebra and get 0.001. This is saying the answer is ANYTHING, but for argument's sake we'll say it 25%. You don't seem to realize that this assumption is not reasonable AT ALL.
I wouldn't survive a day on anything you engineered if you made arbitrary assumptions like this.
what? i only mentioned the approximation thing because i picture segarob as one of those nitpicky douchebags back in college that would ask tons of questions when a problem didnt explicitly state every assumptionTheres a difference between approximating .49 as .5 and assuming that two things that are virtually never equal are equal.