|
PureBytes Links
Trading Reference Links
|
I've seen this particular list go down this alley at least 3 times in the
last 3 years. The last time I invited comment on it asking for any real
honest-to-gawd stats on the walk forward, genetic, IA or NN systems. I
received quite a few emails to me directly and some posted on the chat.
However, none to date really gave any true (not hypothetical or maybe, or "I
think") definitive research with numbers to really show any gain on any of
the above types of systems. I did have one isolated case reported to me but
it was too small to be statistically significant.
Well, I'm still open. I truly would like to see something that is straight
forward and gives a statistical advantage (meaning traded and the numbers to
go with it, i.e. sharp, Max DD, Avg win, Avg Lose, Rina Index, Sharp etc.
I don't expect to see anything usuable come out of this but I sure would
like to. I keep thinking there may be an answer there that will outperform a
simple break out or trend system.
Shall we dance again? Who's open?
Regards,
Michael
*
*>>If the system is back-tested on market data which is similar to our
recent
*>>history, it would most likely provide flat results. The
**>>system needs different market conditions to perform better.**
*
*That is why I advocate a self optimizing code. The system should analyze
its
*own trades and trading decisions against real market conditions and
optimize
*itself to adjust to new market conditions.
*
**A Neural Net working in conjunction with the correct set of data, which
*would need to present to the NN a wide variety of input data configurations
*(about 30 configurations of input data), and most important, providing a
*feedback loop of historical performance versus desired performance, one
*should be able to create a self optimizing system.
*
*Treasure
|