|
PureBytes Links
Trading Reference Links
|
Lou,
you might want to take a look in the AFL library for the Auto-Optimization
framework I did a while ago. It can automatically optimize once right before
trading starts, or regularly every spec'd number of bars, looking back over all
data or a spec'd number of bars. A couple of example functions are provided for
evaluating performance while optimizing, as are some trading systems to test.
<SPAN
class=394482818-01012004>The big problem was that I never found any combination
of systems or optimization parameters that was significantly profitable, and as
far as I know, nobody else who looked at it did either. I was surprised that
using parameter values that did well at point A didn't produce decent
performance going forward after that, but it seems to be true. Are there other
things that could be explored in that direction? You bet; one of the most likely
is other ways besides raw equity growth to evaluate performance while
optimizing.
<SPAN
class=394482818-01012004>
Is
this AI? Not really.
<SPAN
class=394482818-01012004><FONT face=Arial color=#0000ff
size=2><FONT face=Arial
color=#0000ff size=2>
<SPAN
class=394482818-01012004>Dave
<BLOCKQUOTE
>
D. T. and
Other Contributers -
First, let me say thank you for your remarkable imagination
and productivity regarding programs for AMI. I don't post often because
you guys keep me busy studying your posts.
Artificial Intelligence started me thinking about a program
I wrote years ago (days of the TRS-80) which basicly modified a set of
parameters and then went forward and optimized again over a limited number of
days. (Tried a similar program on Race Horses but stocks worked
better) Isn't this basicly what Artificial Intelligence is, modifying or
updating parameters to reflect changing conditions to achieve a final
goal? If so, then a rolling or frequent optimization should give us the
same effect. If so, then determining the parameters to be considered and
giving them appropriate weights would be the first step and then optimizing
the parameters and weights to get the best result. I think you did this
on one of your earlier posts, 20 Systems or 36 Systems? If the above is
true then what we are doing with the current Artificial Intelligence postings
is introducing another parameter, waves and their optimization. I firmly
believe that everything around us is in the form of waves, even the weather
(their prediction is a bit difficult though). As with the
weather, as more parameters are introduced, prediction improves so that I will
be watching you progress with great interest.
A good way to start of the New Year. Best Wishes to
all for a Happy and Prosperous New Year.
lou howard
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
Yahoo! Groups Sponsor
ADVERTISEMENT
Yahoo! Groups Links
To visit your group on the web, go to:http://groups.yahoo.com/group/amibroker/
To unsubscribe from this group, send an email to:amibroker-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
|