|
PureBytes Links
Trading Reference Links
|
The way I understand your problem is this ...
Linear Regression is a 'best-fit' between data points. If you are using
just 2 periods the slope of that line can go from positive to negative (or
vice versa) on the very next bar.
Try extending the periods to something that will not whipsaw you so much.
Peter Gialames
-----Original Message-----
From: owner-metastock@xxxxxxxxxxxxx [mailto:owner-metastock@xxxxxxxxxxxxx]On
Behalf Of Yarroll
Sent: Wednesday, July 25, 2001 7:43 AM
To: metastock@xxxxxxxxxxxxxxxxxx
Subject: LinRegSlope divergence, again
Hello again on the subject.
To sum up your views on the matter, it's the divergence function that's
responsible for the formula weird behavior...
Does it mean that we might as well completely forget the divergence function
because of it? Well, what's the use for it then, (if any)? (I would
understand why some people would use ZigZag function - outside of systems
testing).
It really irks me to say the least. I've done a lot of testing on
divergences... Now it turns out to be rubbish.
All the best,
Yarroll
(I snipped your posts for brevity; here's my original one which Outlook
keeps... hope everyone followed the matter)
I wrote
Hello List,
Please have a look at the formulas:
Buy long: divergence(C,LinRegSlope(C,2),0.01)=-1
Sell short: divergence(C,LinRegSlope(C,2),0.01)=1
They are very profitable in tests, just too good to be true. That's because
it is... Tests results back-adjust themselves (ie, you know you should have
gone long 3 bars after the fact, just like the ZigZag indicator).
My question is: what's wrong with the formula?? For all I know LinRegSlope
indicator doesnt change its apperance on the chart going back. MS Help says
the divergence function uses ZigZag ... is that it?
Thanks,
Yarroll
|