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

Re: Bias in Testing



PureBytes Links

Trading Reference Links

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
> >
>
>
>