|
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/
|