Kelly Criterion for an example

Dforootan

0

Thanks. What's been confounding me is the partial loss here. It seems b is a bet ratio, and partials aren't covered in this version of the formula?

• Kav10

  0

  Win gains 35% and lose loss is 30%. There is no partial loss. See both formulas, they calculate it using two separate approaches.

Dforootan

0

The formula I looking at was f* = [bp − aq] / ab , difference is denominator is multiple loss and gain, that's not accurate?

• Kav10

  0

  No. That’s not!

Dforootan

0

@kav10 i'm going to accept this because the prize is too low. I want to open another question (or I'm happy to tip $50 here) to dig further into this because I've only been able to make all derivatives get the same values if there is a total loss, not a partial loss. You can see https://streetfins.com/the-kelly-criterion-explained/ and https://en.wikipedia.org/wiki/Kelly_criterion (under Investing Formula) f* = [bp − aq] / ab = p/a - q/b and f=(bp-q)/b match when there is a total loss of a

• Kav10

  0

  Looks like you have not accepted yet! I’ve already solved the question with two approaches and answered your questions all for $4. Do you think that’s fair?

• Kav10

  0

  Please state specifically what you need for the additional money. It should take me less than hour as my rate is $60/hour. Or you can just open a new question.

Dforootan

0

Also can see Thorp's page 7, lines 25-28: https://wayback.archive-it.org/all/20090320125959/http://www.edwardothorp.com/sitebuildercontent/sitebuilderfiles/KellyCriterion2007.pdf "One claim was that one can only lose the amount bet so there was no reason to consider the (simple) generalization of this formula to the situation where a unit wager wins b with probability p > 0 and loses a with probability q. Then if the expectation m ≡ bp − aq > 0, f∗ > 0 and f∗ = m/ab."

• Kav10

  0

  Remember that there are two main versions of the Kelly Critereon formula, one used for Gambling and one used for Investment. Regarding your question about partial loss, I did some research and a more general form of Kelly's formula considers the partial loss associated with an investment: {\displaystyle f^{*}={\frac {p}{a}}-{\frac {q}{b}}}

• Dforootan

  -1

  This is not correct. See https://matchmaticians.com/questions/npszwx

Dforootan

0

This = The answer*

• Dforootan

  0

  Per my other question: the correct formula is {\displaystyle f^{*}={\frac {p}{a}}-{\frac {q}{b}}}, where a and b 0

• Dforootan

  0

  comment got cut off … a and b are between 0 and 1. The first formula above in the answer is just the case where a=1 the second formula above in the answer seems to just not work.