Combining multiple systems

Discussion in 'Strategy Building' started by Arrow, Sep 4, 2008.

  1. fvmn

    fvmn

    Because it's that simple :)
    You can try to apply some very advanced math to a coin game, but the damn coin will still land on each side with a 0.5 probability. Not much you can do about this (that's why you shouldn't play coin game or roulette for real money).

    If you do not have a method to predict stock movements you are playing a coin game. Yes, you can apply a lot of math here too, but it won't help you.

    With all the respect to MAESTRO (and I mean this, because I really find a lot of interesting in this posts), this time he is wrong. You can not win in trading just using some math "paradox". Which isn't actually paradoxical after all :)
     
    #121     Sep 24, 2008
  2. I assumed that MAESTRO was not talking about the trivial that frustrated everybody. I solved for the case where there are 4 different continuing series which have 50/50 chance of being heads or tails on the next timeframe. I then assumed that difference between time periods was unknown. Thinking about it now, I'm guessing he meant at time t and t+1, where t could be 1 to infiniti.

    MAESTRO, could you clerify your criteria.
     
    #122     Sep 24, 2008
  3. MAESTRO

    MAESTRO

    Sorry, guys. I am back to the hospital. My health is not letting me to be as active as I wanted. You are all wonderful curious people. If I made you think and try different things I am very happy and somewhat have achieved the purpose. I will clarify a few things when I am back from the treatment.
    Cheers,
    MAESTRO
     
    #123     Sep 25, 2008
  4. mind

    mind

    sh.t. MAESTRO, do well, no idea what your situation is,
    but hospitals are no place you go for vacation. here's
    caring for you ...

    g.
     
    #124     Sep 25, 2008
  5. I wish you well maestro!
     
    #125     Sep 25, 2008
  6. ditto. Hang in there.

    But make sure to get well and come back and keep us all on our toes!
     
    #126     Sep 26, 2008
  7. gbos

    gbos

    Get Well Soon.
     
    #127     Sep 26, 2008
  8. Kerro

    Kerro

    Well, assuming that the probabilities of the stock moving up/down/sideways are 33% at the end of the interval, and the first move of the price is Up, then the prob. are:

    -66% up
    -33% down

    This is with an interval divided into 2 steps. I have not done any further calculations but I suspect that given N steps for the same experiment, one would end up with the fib. ratios TSGann mentioned earlier.

    This sounds very counter intuitive. I think I will be called an idiot in at least one of the next 5 posts for saying this (also considering this is my first post here), but I think Maestro's point was to provide us with the simplest explanation of why markets can be random and trending at the same time.
     
    #128     Oct 1, 2008
  9. Maestro!!!!

    I was playing with some truth tables and by golly, unless I overlooked something, I believe you are correct about looking at the conditional probability in a non-conventional monte hall type approach. I can even see a very possible relevance to parrondo.

    About the only drawback with this type of thinking (as I mentioned earlier) is the assumption of binary betting and payouts.
    Otherwise, it is pretty amazing, and I haven't seen it anywhere else.

    When I get a chance I'll try to simulate some long runs to determine it's veracity, but looking at the implications, I think it is correct and has merit. While I'm not certain I am thinking the same thing you were getting at, looking back at all of your clues and hints, it makes perfect sense.

    Get better. We need more thinkers like you over here!
     
    #129     Oct 5, 2008
  10. mind

    mind

    you got me curious. what did you simulate?
     
    #130     Oct 6, 2008