Fav quotes

[Travis Retriever]

OF COURSE THEY ARE!!! THAT'S THEIR FUCKING JOB!!! It's the definition of being a defense attorney

Well yeah.  Hence why I used it as an example to illustrate Bayesian Analysis.  I was using Bayes' ideas to explain *why* it works that why.  Don't get all pissy with me just because you don't like or aren't comfortable with the idea of living in a probabilistic world.

[dallen68]

Well yeah.  Hence why I used it as an example to illustrate Bayesian Analysis.  I was using Bayes' ideas to explain *why* it works that why.  Don't get all pissy with me just because you don't like or aren't comfortable with the idea of living in a probabilistic world.

Oh, no I'm not  . My drunk ass probably shouldn't have put it that way. (But I still expect tests I take at the doctors office to be accurate.) Other things... not so much. Even on the medical tests, it kinda depends on what it is, though. Like a sonagram doesn't actually tell you have cancer, it tells you you have a growth, and then a biopsy tells you what it is.

[AnCap Dave]

[yt]FLk2hDVTN6g[/yt]

[Travis Retriever]

Fair warning: I haven't read past the item titles in this one, but just looking at those, I suspect it will be full of delicious cluons: http://www.cracked.com/blog/the-6-most-uplifting-facts-about-human-race/

[ArtemisVale]

https://www.facebook.com/pages/Anarchist-Memes/422264631239623

So the page has been "hijacked" by ancaps

Edit: from the Anarcho-Capitalist Bronies page

As some of you may know, A n infinitely hilarious plot was hatched against the now-deceased anarchist memes, what used to be one of the most popular ancom/ansyn pages on Facebook. The page was removed as a result of a report filed way back in June 2013 because the page's admins threatened some other girl on her page.

At least that's what I know, but I honestly don't really care about the reason, I just think it's hilarious that the page was removed and it had over 80,000 likes. Sure, there is a backup page, but it's only at 5,000 likes so far, and it's going to take a while to get those numbers back up.

Meanwhile, us greedy ayncrappers, who don't believe in such silly things as intellectual property, have taken this opportunity to spread our influence. In this page's timeline posts, you will a lot of accepting, tolerant communists displaying their solidarity with their fellow anti-state comrades. Not.

#Falchion

[AnCap Dave]

[yt]DXZIXYM6Q-Y[/yt]

[Travis Retriever]

[yt]DXZIXYM6Q-Y[/yt]

Standing ovation.

[ArtemisVale]

A post from the new anarchist memes i mentioned earlier

"We here at AM don't censor anymore. Not like the past experiences you may have had. So say what you want Comrades....."

Comrades meaning the communist who have been trolling the page lately

[dallen68]

A post from the new anarchist memes i mentioned earlier

"We here at AM don't censor anymore. Not like the past experiences you may have had. So say what you want Comrades....."

Comrades meaning the communist who have been trolling the page lately

I'm sure I understand the problem.

I thought the whole idea of ancaps is the free exchange of ideas. So why not let these statist have their say?

[ArtemisVale]

I'm sure I understand the problem.

I thought the whole idea of ancaps is the free exchange of ideas. So why not let these statist have their say?

Actually they are Ancoms, sorry i wasn't clear. And these ancoms are having their say, thats why I said the page wasn't censoring anyone. Its just that the Ancoms don't have anything meaningful to say and prefer to slander us ancaps.

[Travis Retriever]

"In medicine, you never render a diagnosis based on a particular test result.  To rule out false positives for HIV, your first test needs to be highly accurate and highly precise, which you will get from a lab result, not a test kit. 

If you do turn up positive, you need additional testing such as a Western Blot or PCR method to confirm the positive. 

But if these "gold standard" tests are so good, why not use them first?  The reason is that these tests have a very low probability of a false positive, but are more likely to render a false negative. 

When you combine a low false negative, high false positive test with a high false negative, low false positive test, you get a much more accurate result, especially when combined with clinical findings. 

If Bayesian Inference were taught to med students in their first year, it would make their studies and later careers so much easier.  This is the entire basis for forming a clinical picture of a patient, and they don't even realize it!--Jason Fennec" in the comments of this video (also a win):

[yt]kr5Z-6ub_2k[/yt]

[MrBogosity]

"In medicine, you never render a diagnosis based on a particular test result.  To rule out false positives for HIV, your first test needs to be highly accurate and highly precise, which you will get from a lab result, not a test kit. 

If you do turn up positive, you need additional testing such as a Western Blot or PCR method to confirm the positive. 

But if these "gold standard" tests are so good, why not use them first?  The reason is that these tests have a very low probability of a false positive, but are more likely to render a false negative. 

When you combine a low false negative, high false positive test with a high false negative, low false positive test, you get a much more accurate result, especially when combined with clinical findings. 

If Bayesian Inference were taught to med students in their first year, it would make their studies and later careers so much easier.  This is the entire basis for forming a clinical picture of a patient, and they don't even realize it!--Jason Fennec" in the comments of this video (also a win):

[yt]kr5Z-6ub_2k[/yt]

I decided to try it. For the first test, I assumed that a low false negative was 1%, and a high false positive was 30%. I started with a prior probability of .8%, like in my video.

So I put p(H)=.008, p(D|H)=.99, and p(D|H')=.3 into the calculator and got 0.0259230, which of course became the p(H) for the second test.

For the second, I assumed a high false negative was 35% and a low false positive was 2%. So p(D|H)=.65 and p(D|H')=.02. I put all that in and got 0.4637836213158338. Not quite half.

I don't know what he was considering high and low false positives and negatives, but I'd still want a third test.

[Travis Retriever]

I decided to try it. For the first test, I assumed that a low false negative was 1%, and a high false positive was 30%. I started with a prior probability of .8%, like in my video.

So I put p(H)=.008, p(D|H)=.99, and p(D|H')=.3 into the calculator and got 0.0259230, which of course became the p(H) for the second test.

For the second, I assumed a high false negative was 35% and a low false positive was 2%. So p(D|H)=.65 and p(D|H')=.02. I put all that in and got 0.4637836213158338. Not quite half.

I don't know what he was considering high and low false positives and negatives, but I'd still want a third test.

To be fair, I was mainly posting it for the last bit of his quote more than anything else and possibly the first sentence (pointing out you need more than one test).  The rest was kinda "meh" to me.

[ArtemisVale]

"Capitalism isn't voluntary because your being voluntold to work for me or starve." said some ancom.

I decide to put it in fav quotes because it was so horrible it was hilarious XD

[AnCap Dave]