10$ for upgrading the indicator - page 4

New comment
 
But the snag remains?
;)
 
What is it? That the coefficients are asymmetrical?
 
grell >>:
В чем же? Что коэффициенты несимметричные?

Yeah. Does that say we've moved away from the averaging rule?

Or is it not?

-----

In the example SMA3.3(0)=a1*Close[3] + a2*Close[2] + a2*Close[1] + a2*Close[0], where a2=1/3.3, a1=1-3/3.3;

but maybe SMA3.3(1) should count as a2*Close[4] + a2*Close[3] + a2*Close[2] + a1*Close[1]?
 
avatara >>:

Ага. Это говорит, что мы отошли от правила осреднения?

Или нет?

-----

в примере SMA3.3(0)=a1*Close[3] + a2*Close[2] + a2*Close[1] + a2*Close[0], где а2=1/3.3, а1=1-3/3.3;

но может SMA3.3(1) должно считаться как a2*Close[4] + a2*Close[3] + a2*Close[2] + a1*Close[1]?


It can't.

Let's look at the definition of iMa. The answer is there.
 
grell has proposed a formula showing how the muwings can be continuously transformed from period 3 to period 4. This changes the entire muvings, not their individual coefficients: all the k-values of grell's variant are different from those proposed before.
None of the options proposed earlier are as convincing yet.

avatara, don't be stingy, suggest your geometric option.
 
The formula is only valid for fractional periods, the formula does not work for whole periods. There must still be a solution.
 
What do you mean, it doesn't fit?
 
Knowing periods 3 and 5, the formula does not yield period 4. I checked it twice. Maybe I checked it wrong? Loses fairness in relation to a1+a2=1
 
Well, you don't have to combine 3 and 5 to get 4. There's a formula for 4 that's authentic.
Well, yes, the formula is not universal. OK, we can think about it some more.
 
While we're fussing, Unknow is out there chopping cabbage:)))
12345678
New comment