[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [amibroker] Re: The first new high in 250 bars



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.