Hello,
Nice subject whitneybroach.
I would like to open it to another field, the one of the multi-objective
optimizitation.
Actually random walk, monte carlo can be made "raw", so you extract all
statitstic from each run, and finally you can look for the best solution for
several objective at the same time... but this is too too and... too ...
long time consuming if you have complex system with many parameters.
MPC can be good, but I think it is long time computation too ?
We got IO too (it use neural networks if i am right), thanks Fred, it can
have several goals, but i don't know how it works inside... are all the
goals reached during different optimization process (several
single-objective optimization) or is it real multi-objective optimization ?
Here is a good articles on multi-objective optimization wich review several
methods (for gentics, but same can be used for trading i guess) :
http://www.calresco.org/lucas/pmo.htm
My question is, are their some people here who try to use some of those
another technics of non linear optimisation with multi-objective.
Cheers,
Mich.
----- Original Message -----
From: whitneybroach
To: amibroker@xxxxxxxxxps.com
Sent: Friday, March 16, 2007 5:12 AM
Subject: [amibroker] Detecting data mining bias with modified Monte Carlo
procedure
While reading David Aronson's book _Evidence-based Technical
Analysis_, I stumbled across a modified Monte Carlo permutation
(MCP) procedure that compensates for data mining bias, assuming that
the "best" permutation of rules was not selected with a directed search.
>From Aronson's perspective, this is good news. He views data mining
as a useful procedure in the discovery phase of research. Plus, MCP
does not require out-of-sample data. Thus it is possible to use more
data for mining and still minimize data mining bias in test results.
The likely result: fewer false positives for systems that are
worthless, and fewer false negatives for systems that are valuable.
The paper with discussion and C# code is here:
<http://www.evidencebasedta.com/MonteDoc12.15.06.pdf>.
Aronson's book site, including a link to Amazon, is:
<http://www.evidencebasedta.com>. Separately, I'm looking forward to
the imminent books from Howard
<http://www.quantitativetradingsystems.com/> and Ralph Vince
<http://tinyurl.com/2os2p7>.
Not being a user of IO (or other AB add-ons), I have no idea if this
MCP approach is already being used in the AB community. It looks
interesting to me. MCP appears to require market data and trade data
from every run, not simply the trade data. That suggests to me that
an AB add-on, rather than a completely external program, would be a
more straightforward implementation.
Aronson also refers to a patented boostrap procedure that accomplishes
much the same thing, White's Reality Check, named for Halbert White,
the patent holder. Apparently WRC is not available commercially.
Best,
__________________________________________________________
Découvrez le Blog heroic Fantaisy d'Eragon!
http://eragon-heroic-fantasy.spaces.live.com/