|
PureBytes Links
Trading Reference Links
|
> Cont. contracts or back adjusted contracts are a real turn off, I
> think. T. Stridsman points out the risk of using the "wrong" kind
> of contract oout in his book.
Yeah, and I think his logic is faulty in a number of areas. I agree
that you have to use the right kind of adjustment -- he does all %-
based calculations so obviously he needs %-adjusted contracts. But I
disagree with his fundamental assertion (the value of %-based
calculations) so I tend to disregard a lot of the conclusions he
draws from it.
Example: If running a system on the NDX, Stridsman would have you
use the same calculations (stop sizes, etc) on 9/7/99 as on 12/1/00,
since on both days the price was 2500. But the average daily range
was about 50 on 9/7/99, and 175 on 12/1/00 !! Do *YOU* think the
same size of MM stop would work the same on those two days? I don't.
And in fact I was trading it then so I can guarantee you it didn't.
Basing your stop sizes on price is crazy. You should base them on
current volatility.
> I always felt there is only one way to really test systems and that
> is with the real contract. There is only one software that does
> that and that is BEHOLD!. It does rollovers even with intra day
> data.
And how does BEHOLD! handle indicator calculation? E.g. if I'm using
xaverage(Close, 20) in my system, what happens to that xaverage value
just before & after a roll?
Gary
|