|
PureBytes Links
Trading Reference Links
|
Leslie,
>Perhaps if you rethink your question more in terms of 'what is
the hypothesis that I am attempting to test', you might find
more enlightenment than if you were searching for
"producing more reliable results".
Exactly. The generally consensus is that this kind of optimization is wrong
but I think that the way that you have defined it makes it very acceptable.
The vast majority of people will not do it this way; they'll simply crank up
TS and let it rip with absolutely no rationale other than to find a better
fit. That also is the approach taken by most people in system design. They
data mine without developing a market model.
Seems to me that the strength of TS feeds the human weakness of looking for
the quick, easy way.
Regards,
Mike
|