|
PureBytes Links
Trading Reference Links
|
OK, I know I'm possibly being mischievious here, but I must ask the
following:
Many impassioned arguments break out about Optimization Good vs.
Optimization Evil.
But when I read all the available posts, books, rantings, etc. I get the
sense that the issue is not as black and white as both sides try to make it.
For instance, it's obvious that over optimization is a no brainer danger.
OmniTrader, Deep Insight, and many others are perfect examples IMHO.
It's also obvious that some of the best systems either a) "Optimize"
parameters for each market even if it is over a restricted range, or b) Use
an initial run of "optimization" to deduce the fixed parameters used in a
system that requires no further optimization, and even that is subject to
debate.
So my question is, are we really talking about an "art" more than a
"science" here? And is that so bad? Or is it a matter of simple degree?
This is not to ignore previous posts about the need to mathematically define
the problem or pattern first, but at some point in the process, a variable
is a variable and must be experimented with to deduce a proper value I would
think.
wait... wait... let me blow out this match first before signing off... ;-)
Cheers!
Gene
<DarkMagic>
http://www.UndergroundTrader.com/
|