Author's dialogue. Alexander Smirnov. - page 31

New comment
 

to grasn

Below the thread was someone's idea of a perfectly smooth ZigZag.
I.e. for us smoothness is model smoothness or comparison with a perfect model.

 
Well, yes, in QWMA error is 6 times (should be multiplied by 6). I will specify the post as soon as the information arrives.
( 10 - 15/( N + 2 )- ( 12 - 15/( N + 2 ) = -2 <br / translate="no"> is it a joke?))) as the scales are approximately equal we remove the bracket and obtain -2.
i.e. the coefficients should not depend on N
No, not a joke. Is it possible to subtract them like that? They are not equal - only approximately. But the sum of the k-values for all the bags equals 1, that's for sure. I'll post the proof in Maple shortly.
 

to Mathemat

Reminder: In the second variant of the inductor there is no longer a problem to multiply by 6 (on this stupe from above)
Question: - why do you need a non-smooth SMA in QRMA?

 
And there's nowhere to go without it, Korey. And, generally speaking, the SMA is the smoothest of the three...
 
Mathemat:
And there's nowhere to go without it, Korey. And, generally speaking, the SMA is the smoothest of the three...


this is your 1-quadrat-2 induced FULL Tweak after the P.S. appeared.

 

to Mathemat

First screen PERIOD=8, second screen period=34.
Very springy, mv^2 popolam- da?

 
Love it, I'll finish it properly....
 
Yeah, E=mc^2. By the way, for the first, roughest check of the quadratic regression you can simply put N=3. The curve should run exactly through all the clauses, since the parabola at each given moment is approximated by just the last three bars.
 
to Mathemat
On the subject of energy and smoothness.
What draws attention in your formula is the adaptability.
-The better the movement the more "inflated" (from the original flabby)
The more you move, the smoother your indicator line "inflates" to the shape and size of the nose tip of the blimp.
This peculiarity should definitely be used for poor traders.
Various locksmith methods come to mind, but who better than you ....
 
VBAG:
Well, yes, maths often plays tricks. I won't be surprised, much less argue without knowing. If you have more rational algorithms for calculating mash-ups or regression, it would be very interesting to have a look!

And this LRMA, how will it be with speed?

P.S. This is really long out of the question, I just remembered all of a sudden :)
Files:
movinglr.mq4  5 kb
1...242526272829303132333435363738...44
New comment