|
PureBytes Links
Trading Reference Links
|
On a very simplistic basis, if one were to determine that a 1% change in the
price of say bonds, triggered a buy or sell based on back adusted data and
attempted to use that same 1% on actual data to trigger a buy or sell, the
back test would bear no relationship to actual trading. This is so because
back adjusted data is an attempt to preserve the relationships among
prices---not the absolute prices themselves. Thus one may find that C+1%*C
may be an ideal trigger for back data but the same % may result in excessive
trading realtime.
Regards, Jack.
----- Original Message -----
From: Carroll Slemaker <cslemaker1@xxxxxxxx>
To: jz <jz@xxxxxxxxxxxxxx>; Mel <melsmail@xxxxxxxxxxx>; OmegaList
<Omega-List@xxxxxxxxxx>
Sent: Monday, July 10, 2000 4:38 PM
Subject: Re: Bias in Testing
> I'm afraid I'm missing something here because a 10-point rise in the DJIA
in
> 1940 was a MUCH more significant move than a 10-point rise today. On
the
> other hand, a 10% rise in 1940 is exactly as important to an investor as a
> like rise today.
>
> Also, an N% rise in a raw price series should correspond almost exactly
with
> an N% rise in an adjusted version of the same price series so long as the
> time span of the change is relatively small (small relative to the
lifetime
> of the respective contract during its status as the front contract).
>
> Regards,
> Carroll
>
>
> ----- Original Message -----
> From: "jz" <jz@xxxxxxxxxxxxxx>
> To: "Carroll Slemaker" <cslemaker1@xxxxxxxx>; "Mel"
<melsmail@xxxxxxxxxxx>;
> "OmegaList" <Omega-List@xxxxxxxxxx>
> Sent: Monday, July 10, 2000 1:32 PM
> 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.
>
>
>
>
|