Comments on: Open Thread

By: anonymous

anonymous — Mon, 21 Apr 2008 22:26:12 +0000

Does Bayes’ theorem make racial stereotyping okay?

By: DaveInNYC

DaveInNYC — Wed, 26 Mar 2008 18:28:05 +0000

Quick question on log odds: from what I’ve read, you calculate the log odds in decibels by treating the odds as a fraction (i.e. dividing one “odd” from the other), taking the base 10 log of that, and multiplying by ten. But at http://www.overcomingbias.com/2008/01/0-and-1-are-not.html, it seems that they are calculated using the natural log:

But when you transform to odds ratios, 0.502 and .503 go to 1.008 and 1.012, and 0.9999 and 0.99999 go to 9,999 and 99,999. And when you transform to log odds, 0.502 and 0.503 go to 0.08 decibels and 0.12 decibels, but 0.9999 and 0.99999 go to 92 decibels and 115 decibels.

Earlier in the same post, base 10 seems to be used:

For example, let’s say that the prior probability of a proposition is 0.0001 – this corresponds to a log odds of -40 decibels.

So my question is, am I mistaken as to the definition of log odds, or was this just a mistake in the post?

Also, why multiply by 10? I do not see how that helps out as far as further calculations go.

By: Z. M. Davis

Z. M. Davis — Thu, 13 Mar 2008 01:13:43 +0000

One million hits!

By: Z. M. Davis

Z. M. Davis — Wed, 05 Mar 2008 18:36:56 +0000

Um, mentally edit “command” to read “receive” in that last sentence.

By: Z. M. Davis

Z. M. Davis — Wed, 05 Mar 2008 18:31:17 +0000

Alan, my preferred solution to this problem is to insist on a very strict fact/value, is/ought distinction: ideal rational agents should come to agree on the facts of the matter about how the world is, but they need not value the same things. You can enjoy music more than paintings without making any falsifiable empirical claims.

This is a common view, but by no means does it command universal agreement.

By: Robin Hanson

Robin Hanson — Wed, 05 Mar 2008 18:25:03 +0000

Alan, browse the "disagreement" category of posts, especially this one.

By: Alan Gunn

Alan Gunn — Wed, 05 Mar 2008 18:03:17 +0000

(If you’ve already discussed this, just ignore). Suppose you could get everybody to become completely free of bias, as you perceive it. Would people continue to disagree? To disagree as much as they do now? I ask not because I think these questions have answers, but because I suspect that dealing with them might clarify what you (we?) mean by “bias.”

For instance–I like (some) music a lot, painting, not so much. Is this because I’m “biased” against painting for some reason, or something else? If it’s because of something in the way my brain is put together, would that be a kind of “bias,” or just a legitimate reason for different preferences? If I were blind and not deaf, no one would take my preference as a bias; same if it’s just a less extreme (but still biological) tic?

By: Z. M. Davis

Z. M. Davis — Tue, 04 Mar 2008 19:27:34 +0000

Dmitriy, see e.g., "The Bottom Line," "Fake Justification," and "Fake Optimization Criteria."

By: Manon de Gaillande

Manon de Gaillande — Tue, 04 Mar 2008 19:18:02 +0000

You don’t actually *die* (as far as I know, you can’t reverse death). You are “merely” dying, and the process that leads to death is stopped before you die, until the condition that would kill you can be cured. You may still consider it death, but then you should also regard a person with tuberculosis who doesn’t have access to penicilin as dead. Or do you think a prolonged lack of self-awareness is death? Then you probably don’t sleep much.

By: Dmitriy Kropivnitskiy

Dmitriy Kropivnitskiy — Tue, 04 Mar 2008 19:03:10 +0000

I wonder if someone could point me to any posts in this blog or other sources of information about post-factum rationalization. I am not sure what the official term for this is, but the phenomenon I am talking about is when a person first intuitively and in a lot of cases subconsciously comes up with a position on some issue and only after that comes up with “rational” arguments to justify the intuitive position.