1
$\begingroup$

Is there a absolute dollar form of the Kelly equation $f=\frac{m}{s^2}$? (i.e. one that does not use percent returns).

$\endgroup$
1
$\begingroup$

Well, the first formula on the wiki page gives you a straight forward answer in absolute terms (you do know your bankroll so its pretty much absolute):

http://en.wikipedia.org/wiki/Kelly_criterion

Simple as that, sometimes it does not pay but only causes headaches to overcomplicate things :-)

Happy Thanksgiving!!!

Update as requested by OP:

http://www.math.washington.edu/~morrow/336_10/papers/jane.pdf

and here an application using R

http://braverock.com/brian/R/PerformanceAnalytics/html/KellyRatio.html

$\endgroup$
  • $\begingroup$ That formula seems to be a specific case for binary outcomes, quote: "For simple bets with two outcomes, one involving losing the entire amount bet, and the other involving winning the bet amount multiplied by the pay-off odds". I was more specifically wondering if there was an absolute dollar form in the continuous case. $\endgroup$ – Craig Nov 23 '12 at 19:23
  • $\begingroup$ Added more references for the continuous case and specifically applied to financial assets. $\endgroup$ – Matt Nov 25 '12 at 10:05
1
$\begingroup$

See Ralph Vince's excellentbook: Handbook of Portfolio Mathematics where he goes through explicit, worked examples of using an appropriate modified-Kelly system in dollar / contract terms (Optimal F). He even gives Excel examples for the programmatically uninitiated.

$\endgroup$
  • $\begingroup$ Cheers! That book looks like it's worth a read. $\endgroup$ – Craig Dec 1 '12 at 17:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.