TA or something you don't know about. - page 103

 
Peter_Zabriski:

This is an indicator. Don't make me angry. I am writing his full description.


Interesting fading front. Separate for the up and down pulses.
The start, I suppose, is here.

 

Peter, thanks for the idea.



There's still a fade front to work with. The algorithms are all yours.

 
sv.:

Peter, thank you for the idea.



There's still a fade front to work with. The algorithms are all yours.

What algorithms? Where is the code?
 
Roman.:
What algorithms? Where is the code?
https://www.mql5.com/ru/code/9284
Feed the MACD to the filter input, with separate edges for up and down pulses.
 
О! Thank you. I'll have a look.
 
 

Yes. Sorry. I will write a description of the indicator after all. It's just (making excuses) it's been really hard for me to do. Been - busy.

But I couldn't resist responding to the bullshit. What a load of rubbish!

 

Description.

From afar. Actually, it's strictly... ouch, okay. Let's say I have a market theory, and those who haven't understood it, they have understood something else, and they're cutting the cabbage on it. Really, I don't know anything about it. Which is not surprising. Who, pray tell, will flaunt their abnormality? )))

So. If you look at the timeline of that

 

Here it is. Last subwindow. Description.

No. The principle.

We take the MACD. Detect, i.e. calculate its modulus on a sufficiently long segment. We assume that it is volatility.

Then... We take the MACD peak values, say, 2/12, and postpone them modulo. We obtain a smoothed for several hundreds of bars general idea about volatility.


 
This is a technical layout. Realistically - three or five values should remain. I'll think about it...
Reason: