Oct

26

victorJeff Watson thoughtful post below about the relevance of George Seurat to trading inspired me to think about the many lessons about markets I learned from Sondheim's Sunday in the Park with George, and asked my assistant Linda to send him a copy of it tonight. Two hours later, as I looked through the jacket of my pink coat that I wore to my brother Roy's Halloween party, I found a playbill — Sunday in the Park with George. The 1 in 16 chance of five days' repeating each other like last Friday, makes me think of coincidences and the many times that I have had patterns with 25 of 25 correct predictions at the hand ready for use, only to find the 26th, in real life, totally wrong. The power of probabilities to make truly unlikely events when taken in isolation very probable, a variant of the birthday problem, is astonishing and if one was not a man of lack of faith, it would be eerie.

Michael Cook writes:

This makes me think of Ramsey Theory. The classic version of Ramsey's Theorem: in any collection of six people, either three mutually know each other, or three mutually do not know each other. The philosophy behind Ramsey Theory in general is: "a sufficiently large system, no matter how random, must contain highly organized subsystems."

This is very suggestive for markets.

Steve Humbert replies:

Ramsey Theory is fascinating, but other than as an analogue I'm not sure it has any predictive power in the markets. Note that Ramsey Theory does not reveal which people know each other and which do not (a bit like Ogilvy's famous lament that half of the money spent on advertising was wasted, but that he was never sure which half), and that in its binary, know-don't know, criterion, RT doesn't tell us anything about the strength of the connections. A passing acquaintance is treated the same as a life-long friendship (the market equivalent of treating a 1-tic and a 20-tic up move (or down) as equivalent, and only concerning oneself the binary up-down distinction.


Comments

Name

Email

Website

Speak your mind

4 Comments so far

  1. glenn shaw on October 27, 2008 1:11 am

    Going into Monday's abyss…. tee curves all seeem to go about face ….my monday opens .started with sellin 35 % pos in yen just now… 

  2. Daniel on October 27, 2008 6:33 am

    Re: Binary applications in the market.

    My “guy” took the Hamiltonian Function and applied it to the S&P500. It is unfreaking believable at projecting IT market inflections WAY (months) in advance of the event. As I approach these “windows” I utilize other tools for positioning.

    PS - The secret to the successful application of the H-F is identifying the “Hamiltonian”.

    Happy hunting!

  3. Curmudgeon 5349 on October 28, 2008 10:56 am

    Daniel, your friend is a smart person, but he lives in a world of fantasy, not reality. The Hamiltonian is a perfectly good concept in physics, but it has no bearing on the stock market. Don’t be snowed by fancy words.

  4. vic on October 28, 2008 11:09 am

    one agrees with the curmudgeon not only on this fancy mathematical concept, but all others applied to markets that dont capture the rake or vig implicitly — and that applies to anything in any way related to fourier, chaos, or neural— I could extend the list. vic

Archives

Resources & Links

Search