|
PureBytes Links
Trading Reference Links
|
Hi folks.
Ive spent the last months testing literally dozens
of different systems on different types, and periods, of data. <FONT
face=Arial size=2>I always set the commission to 1% using the settings menu and
the results I get are varied. Some systems give over 100% and others give -100%.
However, taking the best system and changing one
or two variables (all the systems have optimisation parameters) slightly
turns and excellent system into a negative expectancy system. So does using the
same system on a different time period of the same "filter" in the
backtester. The results seem to be random and not system based in that
none produce consistently good or bad
results.
The next step would be to use position sizing and
pyramiding contracts to see if this makes results more constant, but I have a
few questions that perhaps someone can help with. If this info is already
documented anywhere, could someone point the way? Many thanks
...
1) I noticed
someone recently pointed out that using a % for trading costs means that AB uses
that % of equity
2) I also read that AB uses ALL equity
for each trade. Is this true? This is NOT how I would operate so, this
makes the results I get from backtester useless. How can you simulate trading
costs and spread when using position sizing algorithms?
2) If I get two or more signals in the same
direction (a trend) how can I open ANOTHER position? Is this possible to
code?
Thanks.
Yahoo! Groups Sponsor
ADVERTISEMENT
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
|