Patterns and aggregated data as basis for Hidden Markov Models and Bayesian networks

Discussion in 'Data Sets and Feeds' started by Gringinho, Apr 10, 2004.

  1. ...thanks. Alas, today's trade journal provoked little response, proving that it pays better on ET to be outrageous rather than serious.
     
    #31     Apr 12, 2004
  2. hypostomus,
    I will run the companion test as you have suggested. Thank you for your input. I look forward to seeing the results.

    XT
     
    #32     Apr 12, 2004
  3. ...you may get different results for what you trade and for your expenses than I do for mine. I suspect that if you are able to tag highs and lows with a reasonable accuracy that it would chiefly be useful for exiting positions, not entering them. There are lots of people who counter-trend trade here, just ain't my cup of tea for the holding periods I like.
     
    #33     Apr 12, 2004
  4. abogdan

    abogdan

    I think pull-back type of trading could be much more fruitful in this case.
     
    #34     Apr 12, 2004
  5. Hehe, noone said it was trivial stuff. Go to http://citeseer.ist.psu.edu/ for special case research papers on almost everything between heaven and earth; you'd probably add many thousands of equally interesting pages ... :D
    .. and it's all free research information.
     
    #35     Apr 12, 2004
  6. It is one thing not to be interested and another thing to affirm the thing doesn't exist : is it that everything you are not interested in shoudn't exist :D. Who talks about Crystal here : only you :) ?

    Btw did you read the article from Lequeux about intraday pattern (because that's the term used in the article and in fact by scientists when anything deviates from random), since you pretend that you are interested in "psychology" what pseudo pretended psycho explanation do you have for the pattern below ? Veterans of daytrading should know already since this phenomena exists even before 1929 so that I am very amused by "modern" studies discoveries when it is in fact old known things :D.

    <IMG SRC=http://www.elitetrader.com/vb/attachment.php?s=&postid=472883>
     
    #36     Apr 12, 2004
  7. abogdan

    abogdan

    If we are talking about two dimensional sets then any pattern that exists in them could be sufficiently presented by a predicate that is a simple construction of "If - then" statements. Could you be so kind to write one example of such statements so I could understand what are we talking about.
    Cheers,
     
    #37     Apr 12, 2004
  8. From Nassim Taleb
    http://www.derivativesstrategy.com/...1997/1296qa.asp

    "I left Wall Street for the first time in 1991. I was obsessed with price formation. I couldn't understand from the screen how prices were determined. It took me six months to be able to read prices in the pit. Locals basically read information from the order flow and squeezed the weak party. There's always a pack of five or six dominating locals who abruptly change the prices, who bid a lot higher than the previous offer and have the guts to do it, and the rest of them follow. "

    Only a few of floor traders dare to countertrend agressively at the super top or bottom as some of them represent in fact the big money. The rest of scalpers on the floor then imitate them but other little players especially if they are swing traders won't dare to do it so most used strategy by them would be rather to exploit this pattern not by countertrending it but by playing the break of it by making the hypothesis that most probably the high or low has been settled after only a little while according to this pattern.

     
    #38     Apr 12, 2004
  9. For the joker....er...Harrytrader
    "My major current hobby is teasing people who take themselves & the quality of their knowledge too seriously & those who don’t have the guts to sometimes say: I don’t know...." Nassim Nicholas Taleb
     
    #39     Apr 12, 2004
  10. There is nothing special in Renaissance just marketing hype based on the size of their hedge fund as it was also the case for LTCM big size. If you don't remember how LTCM was glorified by nearly everybody except a few ones this is a reminder:

    http://www.derivativesstrategy.com/magazine/bestof/VAR_Chief.asp

    In December 1996, just as the financial world began making value-at-risk a core component of its risk management practice, Nassim Taleb gave a scathing critique of VAR and its defenders. His interview, which was subsequently published on our web site, has probably captured more hits than any other derivatives page. In a debate in our April 1997 issue with Philippe Jorion, a professor of finance at the University of California at Irvine and author of Value at Risk, Taleb refined his position.

    <font color=red>It was clear at the time that Taleb was directing his fiercest criticism at Long-Term Capital Management, then viewed as the ultimate advanced statistical trading firm.</font>

     
    #40     Apr 12, 2004