|
PureBytes Links
Trading Reference Links
|
Mark,
I am wondering if you considered rollover slippage and commission in the
equity curve of your screenshot. From my research I did with WL using simple
methods and probably a similar portfolio money management method this
factors can be very performance eating. Also it would be interesting to see
how the performance was over the last 5/10 years (including slippage and
commission also for rollovers).
Another thing that I found important on my research was the number of
contracts you are able to trade in relation to the volume of the contract.
Finally I wonder how profitable the strategy would be if you take out the
bond (interest related) futures?
Kind regards,
Volker Knapp
(www.wealth-lab.com)
-----Original Message-----
From: Mark Johnson [mailto:janitor@xxxxxxxxxxxx]
Sent: Saturday, October 01, 2005 2:33 AM
To: omega-list@xxxxxxxxxx
Subject: trendchannel, aberration, bollinger bands, etc.
"Volker" <volker@xxxxxxxxxxxxxx> writes:
> Why don't you start by testing a simple standard
> deviation system on a portfolio of futures and use
> simple money management like percent risk stop
I agree. Simple things like standard deviation bands
(a/k/a Bollinger Bands) breakout systems can be
surprisingly powerful.
If you're worried that market selection may be a
dangerous source of "evil curve fitting", you can
sidestep the problem by refusing to perform market
selection. Instead, select ALL markets. Here is
one tiny test of one possible standard deviation
band system, run on a big portfolio that approximates
ALL markets. It's simple yet it makes money.
Other equally simple systems using other equally
simple ideas (things like moving averages, price
channels, etc.) also perform surprisingly well,
when run on very large portfolios with lots of
diversification.
|