Acrary is a genius!

Discussion in 'Strategy Development' started by greaterreturn, May 4, 2008.

Thread Status:
Not open for further replies.
  1. Alan,

    (or anyone else who can answer this)

    You said in another thread that your research in immunological response trading systems changed your perception.

    So you realized using simple systems simultaneously against a market to profit from trend, chop, and counter trend work better.

    You even discussed diversifying across time frames on the same instrument.

    Those points all struck me as genius because I spent several years with the same mis conception.

    It all makes complete and total sense.

    NOW, a logistical question about implementation of this method arises.

    Certainly it can happen that a "trending model" has the account long while a "chop model" triggers a short from a bounce resistance.

    It appears the easiest way to accomplish logistically with my broker is to open 3 accounts. Each one traders a different "theme" against a diverse set of markets and time frames.

    That way each account has a different theme. One is trending, another chop, and another is counter trend.

    Does that make sense or is there a better way?

    Another tips anyone cares to offer are appreciated.

  2. hi Wayne,

    What do you think the net result will be?
  3. You better know exactly how you're gonna manage those positions before, during, and after. What may seem "logical" at first can soon turn into a 3 ring circus with you being the star clown.

    Not saying it can't be done. It most definitely can.
  4. laputa


    rewtree could you tell me the current ET name Alan is using? Pls PM me if needed. Many thanks!
  5. ehsmama


    You ForGot one important system - The one That makes money while markets are closed. Then you have covered all bases.
  6. man


    the only clean solution IMO is to build a software that
    matches orders. thus if one says long, one says short,
    cancel. you need to derive a desired portfolio and place
    only those orders that bring the current to the desired.

    if you do all that "in house" you have the additional
    benefit, that from a certain number of systems in it
    your broker can hardly reverse engineer the overall
    strategy. trades come seemingly at random. or at least
    with very little order.
  7. Man,

    You're absolutely right. I realized it this morning after sleeping on it.

    Why? My broker said I would have to split my capital among the accounts.

    Plus, I realized multiple accounts doesn't solve it really because I also plan to run each "theme" against multi-time frames simultaneously. So the same account (in that scenario) could be both long and short.

    Right, it hit me this morning to make strategies that return a 3 value signal of long, short, or flat. So they internally implement the targets, stops, breakeven, etc and distill it to long, short or flat.

    Then I can apply that same strategy 3 times to the same market it 3 time frames and aggregate the signals.

    long flat flat = long
    long short flat = flat
    long long flat = long X 2


    That would make it infinitely possible to run strategies in a single account.

    Have I got the right idea now?

  8. man


    basically yes. but i would propose to see it even wider. consider
    that one day you would love to make some pyramiding in one
    strategy, or that you do not want to aggregate the same amount
    of capital every day to each system. then you end up with a list
    of markets and a desired position in each. that is going further than
    the concept of longflatflat.

    assume you have three systems and one has a losing streak. you
    do not want to quit, but you don't want to give additional capital
    to it. if you divide your money by 3 every day (which essentially
    happens, when you do longflatflat) you give the losing systems
    more and more allocation on a relative basis, since its own capital
    has shrinked, but it still trades the same number of contracts. now
    if that is part of your plan: okay. but if you do not want to treat
    losing systems like that doing it via longflatflat does not solve the
    problem. but if you recalc your systems, each comes, according
    to its own capital base, with the desired position in each market,
    you cann add up those, compare to current position and again
    trade the difference.

    it is some pain to get this going, but once you have it, you just
    need to add systems and nothing needs to be changed. same
    with capitalinflows: new desired positions minus current equals
    orders to give.
  9. Man,

    That is genius at an even higher level.

    Yes, it is complicated. However, that sounds far simpler than some of the other stuff I have done.

    It may very well take a long time to get it all working but it's nice to know that finally I'm on the right track.

    Fortunately, I have enough experience at mechanical trading to understand the significance of what you say.

    It was very very kind of you to enlighten me, man.

    Okay, now I have a lot of work to do!!

    - Wayne
  10. First of all I don't want to spoil your excitement, but there is something you should be aware of. Your concept is extremely naive and lacks an understanding of the superimposition phenomenon. It is a mind trap that many fall into. The assumption, of course, is hat the equity curve could be smoother by correlating two (three or more) independently behaved strategies. Unfortunately, William Ross Ashby proved in 1961 that it is a dream that has very little to do with the reality. Sorry to bring it to you, but its better be me than markets! In layman’s terms, IT DOES NOT WORK!

    #10     May 5, 2008
Thread Status:
Not open for further replies.