It reduces both risk and reward keeping the ratio steady. It does, however, reduce the min required capital requirements of your portfolio.
MAESTRO you seem to have a special view on distributions. they can by smoother, more uniform, less fat tailed, spiking at the center. which all takes place when you add non correlated systems into one portfolio. yet you insist on the "risk/reward ratio" does not change. you seem to have a very different understanding of the very terms "risk" and "reward" ...
how on earth diversification reduces reward? example: instead of single system with sharpe 1 and 200k allocated to it, we use 2 non-corr systems each with sharpe 1 and we allocate 100k for each, total 200k. returns are the same, but the volatility of returns is lower
In a sense I get the logic you are using but it doesn't account for this fact: It doesn't matter how many 500% 1000% wins you have, if you experience one 100% loss you still end up with nothing.
I missed the part where the OP stated strat diversification increases R:R. He referenced a specific work of Acrary, which doesn't talk about that either. From everything I've read the only one who threw out R:R was Maestro in the misguided attempt to argue something that was never even referenced. All the cybernetics references inclines me to think it's a function of disinfo, but i'll give him the benefit of the doubt and just attribute it to ignorance or poor reading comprehension. Acrary's whole gig was about sharpe, which measures volatility of returns, not R:R. Hard to read and have a discussion about something, if the posters don't know what's even being discussed. It's akin to me starting a discussion on the merits of apples and someone stepping in and creating arguments in counter about the dangers of oranges. Doesn't make a lot of sense.
I probably do have different understanding. It by any means does not make your opinion less valid. I understand where you are coming from and would be delighted to share what I know if it helps you in your trading. Cheers, MAESTRO
I am sorry, but I cannot be clearer. It seems to me that the point I am trying to make escapes you. It is probably my fault and is the function of my limited ability to explain things. I apologize for that. I only hope that by conducting your own experiments the subject would appear more explicit. Please do not hesitate to ask any specific questions as they might help me to deliver my point of view in more rational form. Cheers, MAESTRO
This is how it looks like when two uncorrelated systems are combined. In this case simply added, no Sharpe optimization was done. Alan said the systems should be structurally uncorrelated, which means they are not simply independent but are so structurally different that them being in a drawdown at the same time is little chance. That is very important point. http://www.elitetrader.com/vb/showthread.php?s=&postid=867459&highlight=structurally#post867459
Indrionas, That was very practical and clear of you. I'm working on combining strategies. What software tools do you or Alan use to this type of trading? I'm assuming there's nothing commercially available that has features to intelligently combine diverse strategies. So I'm working on building this myself. But I'd love to hear about any short cuts. Sincerely, Wayne