Rationale for the existence of support lines - page 3

New comment
 
hrenfx:
Whether or not it is considered self-defeating is not an approach. There has to be a method to show for sure that there are no such lines on the SB. And would answer this question in relation to the WTC.
There are such lines on SB (if there weren't, the randomness of straying would be violated). And there are some on the CvR as well. The point is that the lines drawn in the SB cannot be used profitably in practice. But on real market data, perhaps they can be.
 
This is really self-deception, self-deception because it is not enough to predict the turning points, you need to know when such an algorithm will start to lose like hell. But when it starts to fail you will say that you should have traded on the breakout and you will be wrong again.
 
vasya_vasya:
This is really self-deception, self-deception because it is not enough to predict the turning points, you need to know when such an algorithm will start to lose like hell. But when it starts to fail you will say that you should have traded on the breakout and you will be wrong again.
Self-deceivers will pass by.)
 

Most probably this is in the interests of market participants, everyone has his own strategy, everyone makes his forecast with a certain amount of profit (pips), if the forecast comes true they start to think: ok, if it goes this way then it will go this way again. Having drawn the channel, one can make assumptions about the general direction of price movement for a certain period of time and see the most favorable moment to enter the market. If it went in the wrong direction, it turned over (the main thing is not related to the topic, the main thing is to see where the sideways movement will be)) )


SZS: someone clever and big watching you ) (с)

 
vasya_vasya:
This is really self-deception, self-deception because it is not enough to predict the turning points, you need to know when such an algorithm will start to lose a hell of a lot. But when it starts to fail you will say that you should have traded on the breakout and you will be wrong again.


Why is it self-deception? Self-deception is when channels are drawn from close or from high and so on. - The channels are built from close or high, etc. Because the peaks are pleasing to the eye, but the price wasn't really high and that's the point of the peak value? - I build channels at levels where price was lagging because I think the price movement is discrete (even though my question was ignored on the previous page)

http://imglink.ru/pictures/28-10-10/411c90abe63b8c0d8619ed90a08bd589.jpg

 
hrenfx:

The simplest method of drawing support lines that came to mind:

  1. We have the outermost 100 bottom tops.
  2. We start to go through all possible combinations of these nodes and build MNC lines.
  3. Each such line will have a different MNC value.
  4. Obviously the lowest (zero) OLS will be for lines based on two points, so we drop these variants.
  5. We want the line with the lowest Alpha = MPC / F(number of plotted points). Where F(x) = x^n or F(x) = e^x or F(x) = log(x) - we need to see.
Essentially it will be the same regression but for specially selected points (zigzag extremums). Have you got it right?
 

The support-resistance levels are there.

The proof is in the last 2 weeks on EURUSD. Look how briskly price bounces off psychologically important levels - 1.4, 1.39, 1.38.

I have not yet (due to only 2 years of experience) considered the regression levels, although there is room to dig in there as well. But it is obvious that there are at least psychological support and resistance levels.

 
hrenfx:

I don't believe in support lines. You can see them even where they don't exist - SB (random wandering). To see them is to build on history.

Yes it's true, everyone can build their own support or resistance line, you will have one and I will have another, or maybe not one but several.

The problem was voiced by sanyooooook:

it's going down )), it's hitting the upper resistance of the larger channel + considering that 50% of a 3-day move down the probability of a fall is increasing

roughly so

That's the problem, to determine whose resistance is steeper. The channel on M5 or on H4?

In short, it all comes down to the same SB, whether we analyse channels, fractals, zigzags, MAs, etc. etc. All of this, can also come down to the analysis of one pair, but not all. This may work for some pairs and not for others, and even worse than not :) And this may work temporarily or even cyclically.

I think that first of all, we have to learn not to forecast the market (it is useless in the long term), but to work with the market. It's like a puppy that is thrown into the water (it cannot swim) and starts to swim :))

I.e. the main thing is not to make predictions, but to use tactics (God forbid if you think about martingale) to enter and exit the market. but this is IMHO

 
sanyooooook:
It's essentially the same regression, but for specially selected points (zigzag extrema). Got it right?

Yes. The best regression by Alpha criterion is selected from the set of such regressions.
 

Continuation of the same channel:

The price left the channel and entered a subjectively horizontal channel.

It turned out that I drew two parallel lines. After that price reflected from the upper line 6 times with amazing accuracy (even made a profit on it). Also from the bottom line 5 times, also with high accuracy.

In total more than 10 exact points in the future. This, if a coincidence, is much less likely than 10 eagles in a row of coins.

Anyway, it will be up to the research to give the verdict. Again, it's strange why such a study is nowhere to be found. The idea is simple.

12345678910...19
New comment