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

[amibroker] Re: PositionScore Ideas



PureBytes Links

Trading Reference Links

I'm sure you're acquainted with the efficient market hypothesis, 
which argues that it's impossible to make meaningful predictions. Our 
ranking system has proved otherwise. -- Samuel Eisenstadt Of Value 
Line

Whether or not a system is secret, the key issue for anyone 
considering trading a system is how well it has performed on data the 
system has not been optimized on -- that is, out-of-sample data. - 
Dennis Meyers has a doctorate in applied mathematics in engineering. 
He is a member of the Chicago Board Options Exchange (CBOE), a 
private trader, and president of Meyers Analytics. His firm 
specializes in consulting for financial institutions and developing 
publicly available analytical software for traders. He can be reached 
at 312 280-1687, via his Web site at http://www.MeyersAnalytics.com 
or via E-mail at meyersx at MeyersAnalytics.com.

There is a real danger in overoptimizing. When you overoptimize, what 
works in the past doesn't work in the future. When you think of 
artificial intelligence, you tend to think of self-learning and 
changing and optimizing as time goes on. -- David Vomund Of AIQ 

rgds, Pal


--- In amibroker@xxxxxxxxxxxxxxx, "Dave Merrill" <dmerrill@xxxx> 
wrote:
> Let me wade back into the mire by proposing an old idea for 
PositionScore:
> Past performance using the same system being tested. You may say 
that past
> performance is a poor predictor of future gains, but if that's 
really
> fundamentally so, how does the entire backtest itself prove 
anything?
> 
> Simple equity percentage growth is probably not be the best measure 
of
> performance; we've all seen in our backtests how raw CAR gives an 
incomplete
> picture of system performance. So perhaps using CAR/MDD, or Sharpe 
Ratio, or
> Profit Factor, or some other derived performance stat would be best.
> 
> OTOH, while past performance may give some sense of the overall
> profitability of using a strategy on a particular stock, it seems 
to predict
> something only in a longer term sense. It doesn't tell you much 
about the
> likelihood of success of the one particular signal that's occurring 
right
> then. IOW, it's an inherently slow signal, which may be 
inappropriate for
> use as a PositionScore.
> 
> Dave


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->

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 http://docs.yahoo.com/info/terms/