= 27d567bc2d48d9f40e9ccc20ea370b15)
Warrigal comments on The Amanda Knox Test: How an Hour on the Internet Beats a Year in the Courtroom - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (632)
@OP: you have appealed to rationality in examining this case... then you come up with this:
"1. Negligible. No different from the prior, which is dominated by the probability that someone in whatever reference class you would have put Amanda into on January 1, 2007 would commit murder within twelve months. Something on the order of 0.001 at most. "
The FACTS include 1) the police came to "her house" and discovered a murder victim in one bedroom and 2) she was tried and convicted. You seem to have given these zero weighting in your final calculation.
And this:
"3. About as high as the other two numbers are low. 0.999 as a (probably weak) lower bound."
Did your prior of "0.001 at most" apply to Guede as well?
I offer $50 to the AK defense fund if you can produce a defensible Bayesian probabilities calculation showing how you got from your priors to your final probabilities. A condition is that you must account for the fact that most (let us say 80% for the purposes of your calculation) persons convicted of a crime in a democratic society are in fact guilty of it, and that you use generally defensible assumptions.
Should you try to do so, Less Wrong readers can decide if you have succeeded by voting on the post containing your calculation.
D. Alex
While I have a sort of vague sense of disagreement with this comment, I voted it up, because I would be very interested in an example of Bayesian reasoning applied to the real world without having a truckload of given probabilities to work with. In particular, I don't know how one would take into account D_Alex's 80% while also taking into account more specific factors.
I had the same reaction. I'm strongly inclined toward the OP's position, but if you're going to excoriate everyone else for failing to "jettison [their] intuitive feelings in favor of cold, hard, abstract calculation", you should provide the actual cold, hard, abstract calculations supporting your own position.
I should have pointed this out earlier, but for the record: "cold, hard, abstract calculation" referred to the willingness to ignore quantitatively weak evidence even though it "feels important" to you; it did not refer to some specific back-of-the-envelope application of Bayes' Theorem.
(And "excoriate" is definitely not the right word here, at least with regard to the LW community.)