Linear regression channel - page 6
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
I took an object channel and it gives out some wrong widths.
It looks like the standard deviation channel has implemented a strange method of calculating the width, namely the RMS is divided by the cosine of the slope angle. How the slope angle is calculated is unclear. It is supposed to be a coefficient a from the equation y=ax+b, but it will almost always be close to zero, so the cosine will be close to one. So they are artificially increasing it.
That is, with a clear horizontal channel, where the cosine of zero is 1 in any case, their width should clearly correspond to the RMS value.
Why they did that, I don't understand. Actually, I understand it but it is not correct because the RMS is calculated not perpendicular to the channel line but strictly horizontally by bars. As a result, the normal channel starts to widen considerably at a high slope, exceeding the real channel borders.
I don't get it. The middle of the LR is supposed to be the same as the MA. In MT5 there is no such coincidence even with the standard LR.
And the tested indicator does not coincide with the MA as well.
The MA is a polynomial of degree zero, but not a polynomial of degree one.
A polynomial of degree zero:
y=a;
First degree polynomial:
y=ax+b;
The RMS is not calculated perpendicular to the channel line, but strictly horizontally across the bars.
Do you mean vertically? Are you saying that your RMS is calculated perpendicular to the channel line? I don't think there are any such LRs even without acceleration at QB.
The polynomial of degree zero coincides with the regular MA, but in no way the polynomial of degree one.
Forum on trading, automated trading systems and strategy testing
Linear regression channel
fxsaber, 2019.07.28 12:55
The point that is exactly in the middle of the LR segment is the average of the original points. This follows from the definition of LR.
It is at the former that it is obliged to match. Quite on the fingers of the video
It is the former that is bound to match. Completely on the fingers of the video
Polynomial of degree zero, aka arithmetic mean, aka simple waving, aka Bollinger Bond (if you add RMS)
Do you mean vertically? Are you saying that your RMS is calculated perpendicular to the channel line? There doesn't seem to be any such LRs even without acceleration at QB.
Yes, of course I was wrong - vertically.
Mine is calculated vertically of course.
After all, it's not clear how the angle is calculated in the time-price coordinate system.
Polynomial of degree zero, aka simple Mach, aka Bollinger Bond (if you add RMS)
This point must always be on the MA.
This point should always be on the MA.
I don't know about that. It doesn't look like it should. At least not with the built-in tools.
This is the point that should always be on the MA.
No it doesn't. If this point lies on the MA of the same period as the regression channel, then it is generally calculated from the left half of the channel data and the same size data to the left of the channel range. How can it coincide if the calculated data is different?