An interesting game - who can understand from transaction history how an advisor works - page 19

 
YuraZ >> :

that's what I wanted to convey as an idea!

it was with him that several ideas were solved !


So what, you don't want to give the quest to make unraveling that much harder?

 
YuraZ >> :

well logic


channel ... and entry at the pierce ! exit in the middle of the channel

( and the tie not to trade )



To be precise, the output is not in the middle but at the bottom, just on the input and output the width of the channel may be different. This harness is the essence of the EA, without it it would not be an EA, but a moose farm)

 
sanctus >> :

So what, you do not want to give invest to make solving much harder?

>> no I don't because :

manual inputs and outputs are not amenable to mathematical analysis!

therefore there is nothing to figure out!

--

I don't see the point.

--

and i also wanted to say that solving ONLY on the state without password investing is creepy long!

--

your way of trading an automaton because, the point is, there's a clue!

Well, we can see it just fine.

--

how you think the channel doesn't really matter! What's important is that the clue has happened!

 
sanctus >> :

To be precise, the exit is not in the middle but at the bottom, just on the entry and exit the width of the channel may be different. That wrapper is the whole point of an EA, without it, it wouldn't be an EA, it would be a moose farm)

maybe! you can see it on some trades below

I use something like that too.


i tried it on a simple bollinger - i took a flat pair

enter on the breakdown of the target channel in the middle

strange as it may seem

but during testing it turned out that the profit is higher when the exit is in the middle and not on the other side of the channel

 
YuraZ >> :

And I also wanted to say that solving ONLY by steute without a password is creepy long!

You can do it by state too. A suitable script has just recently appeared in the codebase.

 
YuraZ >> :

No, I don't want to because :

manual inputs and outputs are not amenable to mathematical analysis!

therefore there is nothing to figure out!

--

I don't see the point.

--

and i also wanted to say that solving ONLY on the state without password investing is creepy long!

You yourself offered to try and unravel the system. Now you say you don't see the point. Make up your mind.


Manual inputs may not lend themselves to mat analysis, but analysis with the help of the brain may give up)

 
YuraZ >> :

But during testing it turned out that the profit is higher when you exit in the middle of the channel and not on the other side

In my case it turned out that the profit is higher at the bottom exit, but with different channel widths for entry and exit. After optimisation it turned out that the best width for the exit is smaller than for the entry. This is similar to closing in the middle, only it turns out to be slightly more accurate.

 
YuraZ >> :

how you think the channel doesn't really matter anymore! What's important is that understanding happened!


Understanding of the main point has just not happened, i.e. HOW THE FILTER SYSTEM WORKS. With this subsystem, a breakout system, a trend system and any other system will become profitable. I am currently working on the trend system.

 
sanctus >> :

Only IlyaA will know that now, he was the first to figure it out.

Dare I remind you that an EA code was promised to the solver, IlyaA.

 
I wrote that I would probably give the code, and definitely that I would tell the gist of the system.
Reason: