|
PureBytes Links
Trading Reference Links
|
My research has shown exactly that--that is % range or something similar
works best for B/O systems. I find it vastly better than points,% prices,
indicators,etc.--probably due to the self adjusting nature of expanding and
contracting ranges and volatility.
Regards, Jack.
----- Original Message -----
From: M. Simms <prosys@xxxxxxxxxxxxxxxx>
To: jz <jz@xxxxxxxxxxxxxx>; Carroll Slemaker <cslemaker1@xxxxxxxx>; Mel
<melsmail@xxxxxxxxxxx>; OmegaList <Omega-List@xxxxxxxxxx>
Sent: Tuesday, July 11, 2000 12:06 PM
Subject: RE: Bias in Testing
> Why would the percentage "drift" significantly over time ?
> Using POINTS for anything in backtesting is worthless IMHO.
> If a percentage is not appropriate, then what is the best approach here ?
> Using a "% of ATR" approach might be best ?
>
>
> > -----Original Message-----
> > From: jz [mailto:jz@xxxxxxxxxxxxxx]
> > Sent: Monday, July 10, 2000 3:33 PM
> > To: Carroll Slemaker; Mel; OmegaList
> > Subject: Re: Bias in Testing
> >
> >
> > One factor that must be considered when back testing is never to use
> > adjusted price data with systems which use percentages as a trigger. In
> > real time trading, the % factor will not be the same as the back
> > tested one
> > rendering the back test worthless.
> > Regards, Jack.
> > ----- Original Message -----
> > From: Carroll Slemaker <cslemaker1@xxxxxxxx>
> > To: Mel <melsmail@xxxxxxxxxxx>; OmegaList <Omega-List@xxxxxxxxxx>
> > Sent: Monday, July 10, 2000 12:48 PM
> > Subject: Re: Bias in Testing
> >
> >
> > > The degree to which absolute price levels affect backtesting results
can
> > > differ greatly, depending upon the exact nature of the manner in which
> > > performance is measured. Obviously, if simple point gains/losses are
> > used,
> > > absolute price levels will have a huge effect. This effect can be
> > > eliminated by using percentage gains/losses as the measure.
> > >
> > > I'm not sure if this addresses your question or not - it's not
entirely
> > > clear what kind of "bias" you're referring to.
> > >
> > > Regards,
> > > Carroll Slemaker
> > >
> > >
> > > ----- Original Message -----
> > > From: "Mel" <melsmail@xxxxxxxxxxx>
> > > To: "OmegaList" <Omega-List@xxxxxxxxxx>
> > > Sent: Monday, July 10, 2000 3:34 AM
> > > Subject: Bias in Testing
> > >
> > >
> > > > Testing on back adjusted data (going back 18 years) it is
> > not too hard
> > to
> > > > come up with systems that appear to do well particularly in a
> > long term
> > > > rising scenario. Question is how can I have faith in results where
the
> > > early
> > > > back adjusted prices are significantly higher with relation
> > to the real
> > > > prices at the time. Obviously any optimization of any system will
have
> > > > significantly different weighting with relation to the real prices
at
> > the
> > > > time. How can this weighting be equalized - would detrending
> > of the back
> > > > adjusted data reduce/eliminate any bias or should one incorporate a
> > data2
> > > > (real contract) price in the system testing? My goal is to reduce
the
> > > number
> > > > optimization inputs to a minimum so that as a robust system I can
> > > > trust the results. Any thoughts?.
> > > >
> > > >
> > > > Mel Fox
> > > >
> > >
> > >
> > >
> >
> >
>
>
>
>
|