Posts

Help with a (potentially Bayesian) statistics / set theory problem? 2011-11-10T22:30:33.798Z · score: 2 (3 votes)

Comments

Comment by joshkaufman on The noncentral fallacy - the worst argument in the world? · 2013-06-20T17:09:16.884Z · score: 0 (0 votes) · LW · GW

Sorry for the downtime - transferred the domain to a new registrar, and thought the forward would be automatically detected and carried over. It wasn't. Should be back up once the record updates.

Comment by joshkaufman on The noncentral fallacy - the worst argument in the world? · 2012-08-27T20:24:13.830Z · score: 2 (2 votes) · LW · GW

Anyone who visits this page can judge the merits themselves: there's no argument from authority involved. No one is claiming this form of argument is invalid because it's on LW, or because Yvain wrote it, or because it has a catchy name that's published on a website, or because it now has an easy-to-remember URL. I made a simpler citation, nothing more.

Comment by joshkaufman on The noncentral fallacy - the worst argument in the world? · 2012-08-27T14:10:05.916Z · score: 23 (23 votes) · LW · GW

Hahaha, nice.

I was imagining a situation in which someone makes an argument of this type, you say something along the lines of "that's a great example of the 'Worst Argument in the World'," and the person replies "you just made that up..." or "that's just your opinion..."

Providing a pre-existing URL that links to a well-written page created by a third-party is a form of evidence that shifts "Worst Argument in the World" from something that feels like an opinion to the title of a logical fallacy. That can be quite useful in certain circumstances.

Comment by joshkaufman on The noncentral fallacy - the worst argument in the world? · 2012-08-27T06:11:34.699Z · score: 55 (65 votes) · LW · GW

I just registered http://worstargumentintheworld.com - it redirects to this post, and should be available shortly. Much easier to mention in conversation when other people use this argument, and don't believe it's a "real thing."

Great piece of work, Yvain - it's now on my list of all-time favorite LW posts.

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-11T01:24:43.177Z · score: 1 (1 votes) · LW · GW

Wow, that's so simple it could possibly work!

Many thanks - this is most likely what I'll go with. I appreciate your help. :-)

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-11T00:48:34.034Z · score: 0 (0 votes) · LW · GW

Thanks - the voting system analogy didn't occur to me. Reading up on ranked pairs: http://en.wikipedia.org/wiki/Ranked_pairs

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-11T00:40:53.292Z · score: 0 (0 votes) · LW · GW

Ah, I see. Instead of updating half the lists, I was updating the 720 sets where C is the #1 preference. Thanks for the clarification.

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-11T00:00:01.483Z · score: 0 (0 votes) · LW · GW

Very helpful - reading about this now. Starting here: http://en.wikipedia.org/wiki/Ranking

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-10T23:57:38.791Z · score: 0 (0 votes) · LW · GW

Okay, if A is preferred from { A , [B-G] }, that should add probability mass to [A, [B,...,G] ], where [A, [B,...,G] ] is a ranked set of objects where the first slot is most preferred. That would represent 720 (6!) sets out of 5040.

All other sets (7! - 6! = 4,320) should either stay the same probability or have probability mass removed.

Then, the probability of A being "most preferred" = the sum of the probability mass of all 720 sets that have A as the highest ranked member. Likewise for B through G. Highest total probability mass wins.

Am I understanding that correctly?

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-10T23:12:14.687Z · score: 1 (1 votes) · LW · GW

Right - thanks for the correction. Posted a correction in the main text.

Comment by joshkaufman on Help with a (potentially Bayesian) statistics / set theory problem? · 2011-11-10T23:08:01.341Z · score: 1 (1 votes) · LW · GW

Good point about inconsistency... I was thinking that individual responses may be inconsistent, but the aggregated responses of the group might reveal a significant preference.

My first crack at this was to use a simple voting system, where B from {A,B} means +1 votes for B, 0 for A, largest score when all participant votes are tallied wins. What messes that up is participants leaving without completing the entire set, which introduces selection bias, even if the sets are served at random.

Preference ordering / ranking isn't ideal because it takes longer, which may encourage the participant to rationalize. The "pick-one" approach is designed to be fast, which is better for gut reactions when comparing words, photos, etc.

Comment by joshkaufman on The Best Textbooks on Every Subject · 2011-08-10T21:23:01.130Z · score: 6 (6 votes) · LW · GW

Useful clarification. In that case, you should know that the book is currently being used by several undergraduate and graduate business programs as an introductory business textbook.

The book is designed to be a business primer ("an elementary textbook that serves as an introduction to a subject of study"), and business is a very important area of study that rewards rationality. At the time of my original post, no one had recommended a general business text. That's why I mentioned the book in this thread.

I appreciate your distaste for perceived self-promotion: as a long-time LW lurker, my intent was to contribute a resource LW readers might find valuable, nothing more.

If you're interested in the general topic and want a more academic treatment, you may enjoy Bevelin's Seeking Wisdom. I found it a bit disorganized and overly investment-focused, but you may find it's more to your liking.

Comment by joshkaufman on The Best Textbooks on Every Subject · 2011-07-27T05:08:43.462Z · score: 8 (12 votes) · LW · GW

Wow, Duke - that's a bit harsh.

It's true that the book is not densely written or overly technical - it was created for readers who are relatively new to business, and want to understand what's important as quickly as possible.

Not everyone wants what you want, and not everyone values what you value. For most readers, this is the first book they've ever read about how businesses actually operate. The worst thing I could possibly do is write in a way that sounds and feels like a textbook or academic journal.

I don't know you personally, but from the tone of your comment, it sounds like you're trying to signal that you're too sophisticated for the material. That may be true. Even so, categorical and unqualified statements like "terrible" / "cotton candy" / and "little value to offer" do a disservice to people who are in a better position to learn from this material than you are.

That said, I'll repeat my earlier comment: if you've read another solid, comprehensive primer on general business practice, I'd love to hear about it.

Comment by joshkaufman on How to Beat Procrastination · 2011-02-06T17:00:41.339Z · score: 5 (5 votes) · LW · GW

Personally, I use PICS: Positive, Immediate, Concrete, and Specific. Huge improvement when you actually use it to plan real goals.

Comment by joshkaufman on The Best Textbooks on Every Subject · 2011-01-18T20:47:16.113Z · score: 0 (0 votes) · LW · GW

Thanks - glad people are finding it useful.

Comment by joshkaufman on The Best Textbooks on Every Subject · 2011-01-17T20:37:17.015Z · score: 17 (29 votes) · LW · GW

Business: The Personal MBA: Master the Art of Business by Josh Kaufman.

I'm the author, so feel free to discount appropriately. However, the entire reason I wrote this book is because I spent years searching for a comprehensive introductory primer on business practice, and I couldn't find one - so I created it.

Business is a critically important subject for rationalists to learn, but most business books are either overly-narrow, shallow in useful content, or overly self-promotional. I've read thousands of them over the past six years, including textbooks.

Business schools typically fragment the topic into several disciplines, with little attempt to integrate them, so textbooks are usually worse than mainstream business books. It's possible to read business books for years (or graduate from business school) without ever forming a clear understanding of what businesses fundamentally are, or how they actually work.

If you're familiar with Charlie Munger's "mental model" approach to learning, you'll recognize the approach of The Personal MBA - identify and master the set of business-related mental models that will actually help you operate a real business successfully.

Because making good decisions requires rationality, and businesses are created by people, the book spend just as much time on evolutionary psychology, decision-making in the face of uncertainty, and anti-akrasia as it does on traditional business topics like marketing, sales, finance, etc.

Peter Bevelin's Seeking Wisdom is comparable, but extremely dry and overly focused on investment vs. actually running a business. The Munger biography Poor Charlie's Almanack contains some helpful details about Munger's philosophy and approach, but is not comprehensive.

If anyone has read another solid, comprehensive primer on general business practice, I'd love to know.