|
PureBytes Links
Trading Reference Links
|
I think that these thoughts are right on. The stop design and placement are
created after you have a workable model.
Like most forms of insurance or risk control, expect stops to reduce risk
AND return. Hopefully the risk adjusted return improves. Be wary if you
find a poor model becoming a good model because of an artful creation of
stops.
Jim
-----Original Message-----
From: HeyPeter <heypeter@xxxxxxxxxxxxxx>
To: omegalist <omega-list@xxxxxxxxxx>
Date: Tuesday, July 28, 1998 2:22 PM
Subject: Re: Stop Placement Concept Question
>>
>> At 8:18 AM -0400 7/25/98, Steven Buss wrote:
>>
>> <snip>
>>
>> >One reads that trading systems should first be backtested without stops.
I
>> >assume that what is meant by this recommendation is that stops are coded
>> >into the system based on technical indicator concepts (i.e., by the kind
>> >of concepts in #2 above). Am I right in this assumption?
>>
>
>One possible meaning behind this that ideally you want your basic system
>to make money without money management stops first.
>
>In other words, you test to see whether your idea has any predictive
>value first eg if I buy a 4 bar high will the price generally go
>higher? Then apply simple exits like exit after 3 bars. You can think
>of this as the return or reward side of the equation.
>
>After you are satisfied that your basic premise is valid and it has a
>positive return(and not due to curve-fitting etc), then begin to apply
>stops to reduce drawdown, largest loss to make the system more
>tradeable. Think of this as the risk side of the equation.
>
>Usually you will find that adding MM stops will reduce returns but of
>course your risk is being reduced at the same time - classic reward/risk
>tradeoff, no free lunch . . .
>
>Faced with choosing from so many potential inputs, I find starting with
>the return side of the equation a practical solution to system
>development . . .then build and refine from there . . .
>
>My 2 cents
>
>Peter
>
|