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

RE: [investment] RE: [amibroker] Will systems degrade? (was Optimization)



PureBytes Links

Trading Reference Links










Hi Steve –

 

I think you just gave a good example of a profitable
system that did fail.  In my opinion, our models are static
representations of dynamic markets.  Whenever the market characteristics
change in a way that our model does not recognize, for whatever reason, then
that model has failed.

 

But, if you are talking about investing in
MSFT, CSCO, etc because of fundamental data (great earnings, super analyst
recommendation, etc) rather than technical data, I have very strong feelings
that that will never work.  I’ve posted on the HolyGrailSM forum about
that.

 

I wish I could trade yesterday’s
charts – I’ve got some great systems for them.

 

Big grin.

 

Howard

 



-----Original Message-----
From: Steve Almond
[mailto:steve2@xxxxxxxxxxxxxxxxxxxx] 
Sent: Friday, October 17, 2003
12:41 PM
To: amibroker@xxxxxxxxxxxxxxx
Subject: Re: [investment] RE:
[amibroker] Will systems degrade? (was Optimization)

 



I'm not sure this idea of profitable systems 'failing' is actually
true. I (along with millions of others) hit on a superb system in 1998. It was
called invest in MSFT, CSCO, DELL and AOL. (let's call it a momentum system).
It worked superbly until about March 2000. It didn't work after that, but I
don't believe it was due to inefficiencies of the market disappearing. It no
longer worked because the market changed (and the system didn't have the
ability to change with that market). 





NOW I can develop systems that work (with hindsight/data mining/curve
fitting) all the way from 1998 to today. Will they continue to work into the
future? Maybe (although judging just by today, the answer is no....), but I'm
fairly sure they have a much better chance that the MSFT etc. system.





 





Steve







 





<span
>seems like one
corollary of this is that as soon as we figure out something that appears
profitable, we should move quickly, before its advantage disappears. 





 





<span
>it also seems
like the notions of backtesting and optimization are unlikely to succeed, even
more so the further back you go, since whatever advantages they indicate have
probably already evaporated. unless they just happen to be recurring again
now...





 





<span
>frankly, it's
hard to see how rational trading system design is possible in a world like
this. or am I just depressed?





 





<span
>dave









<font size=2
face="Courier New">Send BUG REPORTS to
bugs@xxxxxxxxxxxxx<span
>
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to:
amiquote@xxxxxxxxxxxxxxx 
(Web page: <a
href="">http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: <a
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html


Your use of
Yahoo! Groups is subject to the Yahoo!
Terms of Service. 












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.