The fallacy of "continuous Kelly"

Discussion in 'Risk Management' started by kut2k2, Apr 29, 2014.

  1. kut2k2

    kut2k2

    There would be if any of those market-neutral punters actually had the guts to make a directional trade. :D
     
    #11     May 9, 2014
  2. The problem with Kelly betting is that volatility increases in absolute terms which makes many exit amidst a dip, making them believe its flawed, instead of waiting for a recovery which will be the case if the underlying strategy is really working. So whereas volatility increases, so does returns, leaving risk-adjusted returns unchanged. Nevertheless, due to its aggressive nature, the so called half-Kelly might be a more attractive alternative. In any case, anyone who can run a simulation knows its efficacy.
     
    #12     May 9, 2014
  3. kut2k2

    kut2k2

    It still somewhat amazes me how half-Kelly has half the risk of full Kelly yet retains most (75%) of the expectation of full Kelly. Almost seems magical.
     
    #13     May 9, 2014
  4. ronblack

    ronblack

    Continuous Kelly is leverage, not a risk amount. If you risk 1% on each trade, you can risk 6% by leveraging 6:1. This leverage will grow the equity at an optimal rate. Futures and forex traders do that all the time even if they are unaware of it. Hedge funds use continuous Kelly otherwise they cannot make enough returns after diversifying. The problem comes when those who talk about continuous Kelly do not understand what it means. If CK = 6 it does not mean you must risk 600% of your capital. It means you can leverage your calculated risk 6 times and that may be small enough to be safe. .

    Say thank you Ron. I just want to hear someone say thank you, at least once. Otherwise this world has no future.
     
    #14     May 11, 2014
  5. kut2k2

    kut2k2

    Thank you for what? Posting nonsense?

    Your Kelly fraction is your ideal trading fraction. It already includes a reasoned balance of risk and reward.

    If CK gives a value completely out of wack with your actual Kelly fraction, what possible good is it?

    "Your calculated risk"? What is that, exactly? If it's something separate from your Kelly fraction, perhaps it's something you should reconsider the merit of.

    In the example I posted, the Kelly fraction is 90%. The CK value is 6.65. So tell us: exactly what are we multiplying by 6.65 that won't exceed 0.90 and how did you arrive at this "calculated risk"?
     
    #15     May 11, 2014
  6. I was also confused in the past but these Kellys mean different things (I never used them btw) . The fraction means % to risk on a trade and the other one means % to leverage your purchasing power. To give you an example when you are fully invested like in b&h with no stops like good pop and mom investors %Kelly makes no sense but you could use the continuous case to leverage, i.e borrow money against your positions. I think you are using CK for that then if CK is 6.65 you could borrow 6.65 times but since you cannot do that you will borrow only as much as you can (normally 2x). Note that your CK can drop below 100% in which case you sell and go into cash. As I said I do not trust these things and don't use them but these two you mentioned are not related, %Kelly is for discrete betting with stops and your other you call CK is for continuous investing. I think most MBA courses cover both well.
     
    #16     May 12, 2014
  7. kut2k2

    kut2k2

    So what you're saying is that even though this CK formula is all over numerous trading sites, it is only applicable to long-term investing.

    LOL!

    This clutching to things without fully understanding them is the same lazy mindset that created the Bad Kelly formula.

    No wonder 95% of traders fail.
     
    #17     May 12, 2014
  8. Why are you lolling guy? I brought up a simple example for you to understand because you seem to be ignorant of the difference between discrete and continuous finance. Just tried to help. If you do not want the help it's fine with me. Contact Ernie Chan and he will assist you. He is a nice guy.


    Maybe you should consider the possibility that you do not understand Kelly. Given your claims you definitively do not. Sorry if I tried to explain it to you. Again, contact Ernie Chan and he will help you.
     
    #18     May 12, 2014
  9. kut2k2

    kut2k2

    I wasn't laughing at you, ignorant one. That should have been clear from the context. I was laughing at the fact that many traders have embraced this CK formula as THE Kelly formula they should use for trading.

    As far as me not understanding Kelly, that is side-splitting hilarious. Obviously you haven't read my previous threads on Kelly. More to the point, I have a private formula that is dead on target for traders. In the example above, my private formula gives a value of 0.9064, an error of less than 1%.

    By all means, continue wallowing in the delusion that *you* understand optimal position sizing better than me. I'll laugh all the way to the bank.

    I'm done trying to enlighten mooks in this website, it is the height of casting pearls before swine.
     
    #19     May 12, 2014
  10. A better analogy would be the swine (you) throwing mud around, and representing this mud as the pearls of your wisdom. Educate yourself before you impose yourself as an educator. Clearly, you don't understand what continuous Kelly is, and how to use it.
     
    #20     May 12, 2014