|
PureBytes Links
Trading Reference Links
|
To All,
I really need to write us my results which would be a very comprehensive
report. Anyway simply stated the bulk of the Fussy Logic engine was coded
in C++ as a DLL. My test was based on a simple premise of the action of the
price peaks and valleys and a 25 element truth table with my conjecture what
to do if, for example the peaks were rising but the valleys were flat. The
TradeStation/SuperChart code was rather small. It called the DLL with the
arguments of peak rate and valley rate and the DLL returned a parameter
h_bar which spanned -1 to +1. A buy was generated if h_bar > 0.75 and sell
if < -0.75. Closing the long and short positions occurred at other values.
I could optimize the test cases to find the best h_bar trigger values and
the scaling of the peak/valley parameters within the fuzzy domain. Results
were promising but it suffered form the fact that the h_bar values were not
updated until a new peak or valley occurred. To overcome the drawdowns
between signals I played around with trading the open equity as a stop loss
strategy.
My tests proved that FL can be applied to a trading system. Price peak and
valley are not good parameters.
I hope this helps .... Marlowe
----- Original Message -----
From: Cab Vinton <cvinton@xxxxxxxxxxx>
To: L_Omega <omega-list@xxxxxxxxxx>
Sent: Friday, February 04, 2000 7:28 AM
Subject: RE: Rocket Science
> > FL is a tool that has merit in the decision making as well as control
> > systems. I routinely use a Fuzzy Logic decision method to
> > evaluate trading
> > results. I also developed a FL trading system for SuperCharts.
>
> Marlowe,
>
> I'd be very interested to hear more about the way you use FL.
>
> Which software do you use?
>
> Are there any learning resources (on- or offline) you would particularly
> recommend?
>
> How do you use FL to evaluate your trading results?
>
> Any guidelines on how to go about designing a FL-based trading system?
>
> With Thanks,
>
> Cab Vinton
> cvinton@xxxxxxxxxxx
>
|