Search results

Anybody using or used in the past a Bayesian method in his/her trading? How do/did you like it? If you stopped using it, what was wrong with it? I believe nitro mentioned that he uses Bayesian stops. Anybody else using them? Thanks for feedback.

Bingo. As I said before, even taking winnings out of the game requires surviving a minimum of 34 spins in a row just to get to the point where losing the original account doesn't leave you a net loss. And surviving 34 spins is just barely above 50% likely. Risking a lot for a gain of a little...

Sorry, I thought you were playing the second variation where winnings are taken out of the game. Why would you play the original game? You could lose everything at any spin, including the very first spin. I see you've chosen a ceiling of 50 spins. Actually that's the expected number of spins...

Your p/l is wrong. The most you could win is N*.03 - 1.03 where N is the number of the losing spin.

OK I blew it. You get on average 50 spins before losing your account. So your net gain (on average) is 49*.03 - 1 = .47= 47%

You'd have to win 34 times in a row just to barely equal your account. That works out to almost a 50/50 bet that you'd win your account size before losing your original account. That's not a smart bet, that's a bet for degenerate gamblers.

Not sure if you made a mistake or if this is a trick question. If you meant to say 'you lose your whole account' then don't play the game. Risk of ruin is certain. But if you meant instead 'you lose the whole amount you bet', which is allowed to be less than your whole account, then you should...

For the first two moments, I got radically different results. S1 = 0.0575 ; S2 = 0.043045 ; k1 == s1/s2 = 1.33581136 If we use the alternative mean/variance form, the ratio is even bigger: k1 ≈ s1/(s2 - s1^2) = 1.44695 I don't see how you got 50%.

What are the other two top forums?

Of all the unspecified ways in that insane statement about Obama being worse than Hitler, it's pretty clear that the main one for Scat is that (a) Hitler was White and (b) Obama is Black.

Sig, you must understand that for Scat, everything begins with race. Barack Obama is Black, so Scat is in no way objective about him from square one. Scat's racist posts in the toxic dump known as Politics & Religion are legendary even there. Then Obama had the gall to get elected President of...

Here is my current recommendation for a real-world trading fraction: min[ 0.9/max[ .0001 , 2*max[ -Ri ]_i=1toN ] , ½*( new_Kelly_formula ) ] new_Kelly_formula = max[ 0, s1 ]*( s2*s2 - s1*s3 )/(s2*s2*s2 + s1*s1*s4 - 2*s1*s2*s3) , where S1 = sum[ Ri ]_i=1oN , S2 = sum[ Ri*Ri ]_i=1toN ...

What's the difference between Bayesian updating and the updating the new Kelly formula is already designed to do? http://www.elitetrader.com/et/index.php?threads/a-new-kelly-formula.291307/ Sorry but Wikipedia was no help. I have no idea what "prior" to use.

What distribution? IRL there is no distribution. Just a long and growing string of trade returns. So what do you do?

It's very scientific, it's just not revealed science. And I didn't do it for academia, I did it for me. Screw academia; what they know about trading generally doesn't amount to a hill of beans. What I figured out isn't rocket science, it's half rocket science. So how come the academics...

I disagree. First most Kelly blowups happen precisely because of bad formulae. I didn't write the Bad Kelly thread and the other threads for my health. You're stuck on the fiction that there is some return probability distribution out there that you can reach out and touch, if only you could...

Yes.

This is trivial stuff. Just use a black swan factor (BSF): BSF == 0.9/max[ .0001 , X*max[ -Ri ]_i=1toN ] BSF prevents your next losing trade from wiping you out even if it is X times as large as your current worst loss. Choose the smallest value of X that lets you sleep at night...

S5 and S6 are inputs into the proprietary Kelly formula. Same format as S1, S2, S3 and S4 in the new Kelly formula.

Sorry I misunderstood. Of course the puzzles are total artifices. In the real world we'd just have a long string of trade returns. And that is precisely what the new Kelly formula and proprietary Kelly formula are designed to deal with, although they obviously can deal with the artificial...