Kelly sizing...

Intraday Bill,

Further on this -- the paper I allude to with this information, I do not profit from any more than the author of any paper profits from it;s publication in any journal.

As for the course - I don't take the money it makes (nor do I "give it to charity" but rather send it to the charitable doings I prefer to do with my money, namely, into the commissary accounts of those held in local prisons who don't have anything. But that's MY choice to make)

I understand this doesn't make it free (and there is nothing wrong with free, despite my teasing the list members about this) but I don't do the 2 day course for personal gain. -Ralph Vince
 
Joe,

Using Kelly in most parimutuel sitiuations IS in conformance with what I call the "special case.' Though NOT the optimal fractoin to wager, in the special case it equals the optimal fraction to wager.

The markets are different though, and you should NOT operate under the assumption that kelly is the optimal fraction of your acocunt to risk for a given trading opportunity.

Trading is not gambling. (Think of this way -- if it was the equivalent of gambling, why wouldn't Kelly himself have included it in his paper).

I am NOT trying to disparage Kelly (he;s dead, insofar as I know) just trying to she light and save people the pain I have seen others go through.

-Ralph Vince
 
Joe,

You say "At the very least, you are misrepresenting Kelly either because you don't really understand Kelly or because of your own agenda."

I hope I have made my agenda clear. As for my not understanding (or "misrepresenting") Kelly -- I am trying to propose that it is the practitioners of Kelly who don't understand what it is. It is NOT the optimal fraction to wager to be growth optimal. All I can do beyond that is to refer you to the paper. I would quote the paper, but I am forbidden from that for 6 months (if you go to my website, ralphvince.com, I have other papers on similar topics up -- the reason I don;t have this one up is that I am required NOT to for 6 months from its time of publication.)

I cannot make it any freer or more accessible than that. If I could -- clearly I would.

-Ralph Vince
 
Quote from rvince99:

Joe,

You say "At the very least, you are misrepresenting Kelly either because you don't really understand Kelly or because of your own agenda."

I hope I have made my agenda clear. As for my not understanding (or "misrepresenting") Kelly -- I am trying to propose that it is the practitioners of Kelly who don't understand what it is. It is NOT the optimal fraction to wager to be growth optimal. All I can do beyond that is to refer you to the paper. I would quote the paper, but I am forbidden from that for 6 months (if you go to my website, ralphvince.com, I have other papers on similar topics up -- the reason I don;t have this one up is that I am required NOT to for 6 months from its time of publication.)

I cannot make it any freer or more accessible than that. If I could -- clearly I would.

-Ralph Vince
More...

No need for you to post your paper but I would like an answer to the following in an attempt to better understand where we differ.

I have a set-up where I risk $3.50 to make $1.00 with an 80% success rate. In other words, on average I make $1.00 4 out of 5 times and I lose $3.50 once in every 5 trades. What is the optimal size investment for this set-up?

Joe.
 
Quote from sheepsucker:

Lol, busted!
More...
No kidding. I haven't seen that much ineffectual hand-waving and attempts at distraction since the last "magician" I saw at an amateur talent show. :D

Does RVince assume he's the only guy here competent in basic math? That nobody else here can follow the math behind Kelly and KNOW exactly what went on there? :eek:
 
Quote from rvince99:

Ok.

Go figure it out yourself then.

-Ralph Vince
More...

Ralph, I don't want to assume anything. Is this your response to my question below?

I have a set-up where I risk $3.50 to make $1.00 with an 80% success rate. In other words, on average I make $1.00 4 out of 5 times and I lose $3.50 once in every 5 trades. What is the optimal size investment for this set-up?

Joe.
 
Quote from rvince:

Along these lines, the Kelly Criterion is not necessarily applicable trading because, contrary to what Kelly presented (p925) in his paper, it does not yield the optimal "fr action" to risk, but rather, a risk factor. That's ok, provided you understand that going in -- if not, you will be likely wildly overly-aggressive. The Optimal f formula DOES provide the optimal fraction.
It's possible to translate between the two, just as one might translate between Fahrenheit and Celsius. There is an enormous benefit to using the optimal fraction, however, as it is possible from that to determine what the best guess for the future value for f (as a fraction) will be.
I won't elaborate further on that, but if you;re interested this is detailed in the forthcoming Journal of the International Federation of Technical Analysts, presented in Berlin next month, or in Tampa in November (see ralphvince.com for more on this) -Ralph Vince
More...

Can f ever be larger than Kelly? The reason I ask is because that's the answer I got from an online optimal f calculator. I realize you're not responsible for everybody who sets up an online calculator and the results, but I'm curious.

Right now, my Kelly is about 35%, but I'm getting an optimal f of .75 on that same trade series and if you run the trade series as a set of Kelly outcomes (recalculating Kelly after each trade), the optimal f result is .43, also larger than the Kelly number.

Again, realizing these results could be an outcome of a bad implementation in a calculator, but does the theory allow for such outcomes? From what you are saying in this thread, it should not and that, at most, my Kelly and optimal f could be equal.

If you know of an online calculator you know is legit, can you post a link?
 
Hey guys, I also don't like R Vince's refusal to answer specific questions a couple of posters asked but I can understand his points and where he is coming from. He is basically saying amongst other things that Kelly can cause ruin. This is true because Kelly involves stochastic components. R. Vince essentially agrees with Michael Harris in that one. Harris also states in his paper that Kelly can only apply to systems with stationary values for p and R:R. Here is the link to that free paper link.

If one has developed a new system the question is when does one start applying Kelly? You do not know whether the system will remain profitable or generate a large drawdown and lose money. If one applies Kelly to an unprofitable system, then that guarantees ruin. Some other method must be applied until one is sure about the system parameter values and especially the win rate p and the R:R ratio.

I therefore think that R Vince is correct to advice against Kelly because most are struggling with their systems. Maybe some here have systems with stationary values that are profitable but most do not.
 
You must log in or register to reply here.
Back
Top