[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Optimal f and system optimization


  • To: metastock@xxxxxxxxxxxxx
  • Subject: Re: Optimal f and system optimization
  • From: rst@xxxxxxxxxxx (rudolf stricker)
  • Date: Sat, 7 Aug 1999 09:06:39 -0700
  • In-reply-to: <37919bc4.8775155@xxxxxxxxxxx>

PureBytes Links

Trading Reference Links

On Wed, 28 Jul 1999 08:47:12 +0200, you wrote:

>How do you test the system optimization? Did you apply the results to "future"
>data? Do you check how the parameters of the optimizacion evolve with time? I
>have made something in this line (just one system -cross of the price with a
>mean-) and I was very surprised with the results...
>To make this type of complex optimizations, I think the only way is to write
>our own program. I have made my calculations with a program writen in fortran
>(coupled with excel)

Because imo (nearly) all successful systems & beings on earth are
"adaptive", I also use an adaptive trading system, which is optimized
e.g. every weekend to "extrapolate" the results for the next week. The
model parameters show some changes vs time, where the amount of
changes imo reflects the changes in "market behavior". 
To strengthen  the adaptiveness of the system, the profit in my goal
function GF is exponentially weighted vs time to enforce the influence
of "new" market behavior.

mfg rudolf stricker
| Disclaimer: The views of this user are strictly his own.