|
PureBytes Links
Trading Reference Links
|
>What you are experiencing is positive serial correlation in trade returns.
ie.
win begets win, loss begets loss. You could probably improve things even more
buy loading up after wins.
Unfortunately, I can't, so I haven't bothered looking at this. I'm trading
mutual funds, and margin/leverage is very restricted. For all practical
purposes, I can trade with 0%-100% of capital, not more.
>There are lots of articles in TASC and Futures about running stat tests to
determine your 'z-score' to measure the serial correlation.
I've read them, of course. But didn't really understand the math. So I
figured that I'd start with a simple experiment so that I could begin to see
for myself what the dimensions of the problem are. Then I figured it'd make a
heckuva lot more sense when I read what people had done to solve them.
>However, strange as it may seem, you cannot fully exploit a market system
until you have removed the serial correlation. Because after you have remove
the s.c. you can use fixed fractional strategies to really jack up your equity
growth. Like optimal f. But these tools require that the distribution of
returns not be serial correlated (more random). See Vince's excellent work.
I agree. That sounds pretty strange, and completely the opposite of what I
was
looking for. Will I be able to take advantage of these strategies without
leverage? If so, sounds like it's time for me to buy Ralph's book.
-Paul A.
|