Money Management

Discussion in 'Risk Management' started by cnms2, Nov 22, 2005.

  1. Apparently we practically would have to take so many factors into consideration to determine a proper size, and several of these factors you mentioned above have not been included normally with the Standard Kelly formula. Some factors would be even allowed to choose certain subjective values. Every trader might use different number of factors. There can be also different ways and calculations, besides yours, to find out the desired fraction, probably having different results against yours. ... ... ...

    If anyone still calls this overall approach a Kelly formula/ value, that's fine with me.
     
    #171     Dec 20, 2005


  2. Q

    http://www.elitetrader.com/vb/showthread.php?s=&postid=930367#post930367

    UQ
     
    #172     Dec 20, 2005
  3. gbos

    gbos

    The theory behind the Kelly approach is one. Same inputs will result the same output. The question ‘which system is better’ will be answered with the same answer by anyone. The question ‘what is the best possible allocation between a number of correlated systems’ will be answered with the same answer by anyone also.

    The difference is in the personal preferences of risk. Different people will follow different fractions because of this reason and not because of different calculations. It’s a tradeoff between how quick you want to achieve your goal and how smooth ride you want to have.
     
    #173     Dec 20, 2005
  4. cnms2

    cnms2

    This is an upgraded version of the The risk of never being unhappy table.

    <img src=http://www.elitetrader.com/vb/attachment.php?s=&postid=931029>
     
    #174     Dec 20, 2005
  5. Thanks for your feedback.

    There are Two Kelly formulas: Original per Kelly's original paper; Simplified as mentioned many times in this thread. Does these Two formulas produce exactly the same value? Which One you use so far?

    One calculation can base on theoretical probability values, another historical values. Which One you use so far? If base on theoretical values, the results will be far from practical. If base on historical values, the results wouldn't be the same each time with new data. Which One you use so far?

    ... ... ...

    Telling every trader simply to use "Kelly" won't be enough.


    If at least One element with subjective value is allowed in a calculation cycle for a formula/ approach, should we say the whole approach can be used to derive objective results? Using Kelly usually is to expect an optimal value for optimal results, allowing trade-off as personal preference is hardly optimal, imo. ... ... ...

    "The theory behind the Kelly approach is one." - I agree. That's probability theory.
     
    #175     Dec 20, 2005
  6. Surely you know what you're talking about. Keep undating "The risk of Ever being unhappy"! :D
     
    #176     Dec 20, 2005
  7. cnms2

    cnms2

    Actually: "The risk of nEver being unHappy" ...

    There's no sure thing in trading as in life, but probabilities suffice in most situations.

    You may remember the Stone's:
    • "You can’t always get what you want
      But if you try sometimes you just might find
      You get what you need"
     
    #177     Dec 20, 2005
  8. cnms2

    cnms2

    Firstly: the input is subjective, the formula is objective.

    Secondly: the Kelly criterion is not a probability formula, as it is not related to any probabilistic distribution; one of its "subjective" inputs is the probability of winning.

    The Kelly ratio is not a theoretical approach as "1+1=2" in not a theoretical result either ... :)

     
    #178     Dec 20, 2005
  9. #179     Dec 20, 2005
  10. cnms2

    cnms2

    OddTrader,

    I don't get you: do you try to be funny? are you arrogant? are you trying to understand money management but you can't? you understand it but disagree?

    Please explain. If you think that if I work a real life example from A to Z will be worthy to you, I'll consider spending my time doing it.

    Again: I don't get your posts. It's probably me ...
     
    #180     Dec 20, 2005