Channel, what are you? - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Or maybe the channel should be made with tape (flexible) at once a la Bollinger or envelope? avatara, you said something about the non-thickness of the difference between the muv and the price - or did I misunderstand you...
in the case of a bending ribbon there is the problem of predicting its future position, which in the case of a straight ribbon is inherently inherent
А, может, канал лучше сразу сделать лентой (гибкой) а ля Боллинджер или конверт? avatara, ты ж ведь что-то говорил о нетолстохвостости разницы между мувом и ценой - или я тебя не так понял...
P.S. Мечта идиота - сделать осциллу без эффекта насыщения вблизи краев.
The ribbon can be a trajectory of the end points of the channel, but the regression part does not change, at a specific point we use the channel construction and make a trading decision.
I mean we can build a channel at any point (in every point it will be different), but the build method is the same.
Это там, где он о чужих автомобилях говорит?
off topic. enlighten me!
I mean figure 5.2. - pipe ... double logarithm
No, there's also a lecture with his drawings, I can't find them now. It seems to show the inputs, too. If I find it, I'll send you the link.
А, может, канал лучше сразу сделать лентой (гибкой) а ля Боллинджер или конверт? avatara, ты ж ведь что-то говорил о нетолстохвостости разницы между мувом и ценой - или я тебя не так понял...
P.S. Мечта идиота - сделать осциллу без эффекта насыщения вблизи краев, но чтоб края были фиксированными.
And catastrophes - should they give kinks? That is, I personally consider "bloating" of the ribbon at the kinks as "unphysical".
An oscillator without a saturation effect near the edges will have a black hole near zero :)
А катастрофы - они-то должны давать изломы? То есть "раздувание" ленты в местах изломов лично я считаю "нефизичным".
Yeah, who can argue with that?
And now a question for you: is support always parallel to resistance in a channel?
The question is whether the process is common to both support and resistance (then the lines are always parallel) or if it is separate processes that form support and resistance separately then the channel may well be a triangle,
or as separate processes forming support and resistance, in which case the channel may well be a triangle.
If the bulls and bears are different (and there seems to be no doubt about it) then clearly the formation of support and resistance are different processes. But even if we forget that, the parallelism of the channel boundaries is not dictated by anything. It's just a linear regression is built and the borders are drawn from it as from the central line. Hence the parallelism.
IMHO, I've always built the channel separately: top rail by zigzag tops, bottom rail by bottoms. This automatically makes horizontal channels as well as ascending/descending ones. And parallelism is observed only in exceptional cases. Although substantial non-parallelism is also quite rare.
1. first, the brain, analysing graphical data (I'm skipping the step of converting light impulses on the retina into meaningful quotes:), chooses a scale for analysis.
...
2. out of all areas for further analysis, the brain chooses those which are well enough approximated by a linear dependence on time.
...
3. And now for the hardest part. At "quite linear" segments our brain plays a trick - the price must behave "quite regularly".
...
Interesting topic. I would like to find an answer myself - I want to implement the channel strategy in EA. But how do I explain to the machine that here we have a channel, and here we do not...
1. Yes, scale determines a lot.
2. A channel might not be linear - say parabolic, and my brain still picks it out. But no one prevents you from imagining such a channel as a sequence of linear channels with different characteristics.
3. Imagine a sequence of fairly smooth calm trends lasting 3-5 days. Several such trends will be identified as a channel (according to your condition), but the mechanism of their formation is quite different. So, point 3 is not quite suitable IMHO.
The channel defines the fluctuations of the market price between the states of shortage of demand and supply within the average price trend. We need to clearly define what we call a trend and what fluctuates around the trend. If, for example, 3 days is already a trend (TS is working with short positions), then the period of price fluctuation in the channel should be significantly less than 3 days. This is essentially a disclosure of the scale issue. Also it is necessary to distinguish the presence or absence of the average price trend (analog of a trend). This can be identified using a linear regression (if we are talking about a linear channel). If there is no trend, there is no reason for a channel to form.
PS: Why have you abandoned the subject? Although you solved the channel construction problem more than once, have you solved the channel presence/absence formalisation problem? If we use standard methods to build a channel where there is none, we will get nothing. Please send me the link if this topic has been covered elsewhere ... I didn't find it on this forum. I did not find it on this forum.
lsu:
It is an interesting topic. I would like to find an answer myself - I want to put a channel strategy into EA. But how do I explain to the machine that there is a channel here and there isn't...
1. Yes, scale determines a lot.
2. A channel can be non-linear - say parabolic - and the brain still picks it out. But no one prevents such a channel from being represented as a sequence of linear channels with different characteristics.
3. Imagine a sequence of fairly smooth calm trends of 3-5 days duration. Several of these trends will be identified as a channel (according to your condition), but the mechanism of their formation is quite different. So clause 3 is not quite suitable IMHO.