From theory to practice - page 1247

 

A question of ingenuity.

Here's a polynomial regression curve on any graph.

Whatever graph you give it. It'll fit on any graph.

Question. How can we tell if the graph is a channel?

We took a chart and applied a regression to it. What characteristics should we look for in order to clearly state that the graph is or is not a channel?

 
multiplicator:

Question for cleverness.

Here is a polynomial regression curve on any graph.

Whatever graph you give it. Whichever graph you give it.

Question. How can we tell if the graph is a channel?

We took a chart and applied a regression to it. What characteristics should we look for in order to clearly state that the graph is or is not a channel?

A channel can be identified even without regression.

There are several channels on any graph at the same time.

It depends on the time interval under study.

 
multiplicator:

A question of wit.

...

Tell me some indicators that can be calculated.
Maybe look at the variance there. Maybe build a distribution. Maybe the channel has some kind of distribution and not the channel has another kind.

At least tell me graphically how you see the channel. What properties does it have. To try to transfer this information to machine code.
 
Uladzimir Izerski:

A channel can be identified even without regression.

There are several channels on any graph at the same time.

It depends on the time interval being analysed.

how? can you describe how by eye you see it to write it into a programme?
 
multiplicator:
how? can you describe how by eye you see it to write it into the program?

Here is a drawing of the channels. Constructed by a machine without regression. Closer to regression, but better for the job.

1_GBPJPYM5

 
Uladzimir Izerski:

Here is a drawing of the channels. Constructed by a machine without regression. Closer to the regression ones, but better for the job.


Well still some indicator is used, if not regression.

so how do you determine the beginning of the trading channel? ah, it's a company secret)
 
multiplicator:
Well, anyway, some indicator is used, if not regression.

so how do you determine the beginning of a trading channel? ah, it's a company secret)

No, it's not a secret. On top of the previous wave.

Waves and channels are inseparable.

 
multiplicator:
Well, some kind of indicator is used anyway, if not regression.

so how do you determine the beginning of a trading channel? ah , it's a firm secret)

The beginning of such a channel can only be determined ex post facto. It usually starts from local extrema.

 

Here is the continuation of the unshaded blue channel. Red has worked out and turned into support.

Even changed the pattern. It's even more interesting.

2_GBPJPYM5

 
Renat Akhtyamov:

Chehegewara, people here are still in space...

Tell them.

;)

There's no time to explain)))
In general, for example I already know and use three types of trading. Each of them works out by 80%-85%. But all this is based on chance and people losing their hard-earned money.)
There are no secrets. As always, everything brilliant is simple:)
Reason: