|
PureBytes Links
Trading Reference Links
|
I thought about that scenario.....I guess it indicates the COMPONENTS of our
so-called, made-up index were perfectly inversely correlated......
this would tell us we COULD NOT generate or optimize the system for BOTH
components in combination....there would be little or no trading !!!!
However, it would make for a terrific options trading strategy......sell
puts on the strong component and sell calls on the weak one.
> -----Original Message-----
> From: Phil Lane [mailto:patterntrader@xxxxxxxxxx]
> Sent: Friday, November 12, 1999 11:58 AM
> To: M. Simms; Mark Johnson; omega-list@xxxxxxxxxx
> Subject: Re: Trying to optimise beacktesting but TS doesnt have feature
>
>
>
> ----- Original Message -----
> From: M. Simms <prosys@xxxxxxxxxxxxxxxx>
> To: Mark Johnson <janitor@xxxxxxxxxxxx>; <omega-list@xxxxxxxxxx>
> Sent: Friday, November 12, 1999 8:37 AM
> Subject: RE: Trying to optimise beacktesting but TS doesnt have feature
>
>
> > I understand the problem.......there's got to be a way around this !!!
> >
> > Would it be appropriate to generate your own data series using the
> combined
> > price-relatives of the symbols you want to optimize together.......
> >
> > if OEX is up 1.2% and SOX up 1.0%, then the combined move would be
> 1.1%....
> >
> > Think about it......any theoreticians out there ???
> >
>
> this might be a good idea for stocks. You'd wind up filtering out
> the noise
> in individual stocks. Should wind up being equivalent to running
> the system
> on a broad stock index - ?
>
> I've had "some" good results combining the various stock indexes, and/or
> using one against the other as a confirmation. But - what if the
> markets you
> were combining were exactly opposite of each other? Your system would be
> running on a flat-line chart. hmmmm..
>
>
>
>
|