Geometric approach in price forecasting - page 3

New comment
 
Martingeil:

According torobi11's terms of reference.

On Mikula's theory.

We take the maximum1.4939(EUR/USD, daily), remove the comma (or, in mathematical terms, multiply by1000), obtain 1493.9, round up to the whole number and obtain1494.

We need to find the support levels.

For this purpose we take the root of1494, we get38.652,subtract(because we are dealing with a price peak) the increment 0.25, we get38.402, squared, we get1474.7, we multiply by1000 and get 1.4747. This will be the first support level. And so on.

38,652 - 0,25 = 38,402^2 = 1,4747

38,652 - 0,333 = 38,319^2 = 1.4683

Increases:

Found Mikula's Theorem turns out to have been, but forgets everything. What do you think the author about it ...

Damn, I made a mistake, where is this robi11 now? What do I do next with the total, how do I turn it back? - $230.43 ???

I read Mikula, it's 230, 43 coordinates of square 9, something's wrong here, he's not telling the truth, robi11, you shaitan!

Good day!

I read this Mikulu, as well as other works, including those by Gunn (and almost in the original, translated by Hierzik, etc.) ...

What is the point of using the square 9 (as well as 4, 12, 24 and other heresies)? It's a definition of support/resistance levels! How is it done? Correctly you wrote "... We take the maximum ...", i.e. the calculation is performed from some extremum, which is kind of a 100% reversal point. Then it is calculated in what corner of the square the level is located and other near diagonals are identified. The system proposed by Yeremeev works in a similar way. It also calculates levels starting from an extremum, but with the use of coefficients determined using the price motion chart of a specific symbol (I can give more details for those who are interested).

But here is the main question - how do you know that this reference extremum is a global one? I'm not so sure about it!

The method I suggest is good as its prediction for a local movement - a possible reversal (just a local one) appears, we have calculated the first impulse, predicted a possible movement - the movement has worked out, that's all! We are further waiting for a possible reversal. Each movement is unique! And the forecast is being made and corrected in the process of price movement - right here, right now.

Now about the theory of Gann: He's determined everything normally, his approaches to trading are unique in their kind, but they are outdated and they should be adapted to reality - I can even tell a little about it, I've even written a recent post about the fluctuation of prices and the use of the simplest neural network to predict this very fluctuation. By the way, just using his approach to identify small, intermediate and major trends helped me to avoid abstractness of determining the reversal using the method #2 I haven't told you about yet ... :):):)


Regards, RomFil

 
Martin CHEguevara:
I've been trading in the market for quite a long time... And I know everything there is to know about market movements in order to make money...
I can reduce any absolute system to the simple principles by which it works.
In your case, it is a time-dependent pivot. Your charting system is based on the property
Prices form a flat in a natural way by reducing the intensity of max/minimum crossing or, as in your case, by increasing the time of crossing the extremums.
The reasoning and discussion is very long. But I think everything should be clear.
I'm not trying to criticize you. I am only pointing out your shortcomings. For a successful person, this is a reason to get better;)
I don't claim to be unique in the method I'm proposing. I only found a rather simple way of interpreting (at least it seems so to me :)) exactly what you wrote about. Thank you for the shortcomings of course, but it's impossible to point out the shortcomings of a method until it's presented ... :) Or are you a psychic? Perhaps of course you already know about what I want to set out, because this site is the second resource where I have laid out the approaches I describe.
 
Taras Vavryn:

Yes it's good to write everything down when the chart is already drawn, do it at the first tick starting around tomorrow!!!

Enclosed screenshot "Special for you" ... :)

I entered the trade last night - I entered on high time. Today at 4:55 terminal time a signal appeared, which says that the price will go up. Then I built up targets according to the method. At 7:15 I got the signal to go down. I have built the goals and as the price moves, I have seen a certain resonance zone appeared and the price is very likely not to break through it but bounce from it, although it will not reach the downside target quite a bit... It is like the LPS signal from the VSA methodology... :) Now we have to watch the price. How it will touch the lines and rebuild the targets on the fact of touching. Something like this ... :)

Files:
flr5.png  34 kb
 
In general, for more than half a year I have had the obsessive idea to define / predict targets / levels with only the first momentum. Unfortunately, your method has awakened the idea again, but the difficulty is in the charting itself. In fact you are able to define future levels using the existing momentum, but here the question hangs on. Predicting the level in your case takes price and time into account. I do not think that time prediction should be as strictly consistent as price prediction. But intuitively, for some reason I think it is the other way round.
 
RomFil:

The attached screenshot is "Special for you" ... :)

I entered the trade last night. Today at 4:55 terminal time a signal appeared, which said that the price will go up. Then I built upward targets according to the method. At 7:15 I got a signal to go down. I have built the goals and as the price moves, I have seen a certain resonance zone appear and the price is very likely not to break through it but bounce from it, although it will not reach the downside target quite a bit... It is like the LPS signal from the VSA methodology... :) Now we have to watch the price. How it will touch the lines and rebuild the targets on the fact of touching. Something like this ... :)

Interesting, we'll see. But it seems for the first (left) trend target is not defined correctly, the midpoint should be on the previous bar.
 
Anatolii Zainchkovskii:
I've been obsessing for more than half a year over the idea of setting / predicting goals / levels using only the first impulse. Unfortunately, your method once again awakened the idea, but the difficulty is in the graphical construction. In fact, you can define future levels using the existing momentum, but the question hangs there. Predicting the level in your case takes price and time into account. I do not think that time prediction should be as strictly consistent as price prediction. But intuitively I think it is vice versa.

Why unfortunately? :)

If you are looking for something it means you haven't found what you want yet... Maybe that's what you're looking for... :)

To be honest, I don't understand the meaning of the phrase: "Alas, the time values for the price forecast don't correspond with ..." This method sets only goals (levels in another way). About temporal forecasts - it's method number 2. I will most likely tell you about it today - a little later.

 
RomFil:

Why unfortunately? :)

If you are looking for something it means you haven't found what you want yet... Maybe that's what you're looking for... :)

To be honest, I don't understand the meaning of the phrase: "But the time-based price forecasts, alas, no longer correspond to ..." This method defines only the objectives (levels in another way). About temporal forecasts - it's method number 2. I will most likely tell you about it today - a little later.

I think that temporary forecasts should go along with price forecasts. And if the price forecast is not in line with the time forecast then I think it is a fluke, while the true forecast should confirm both price and time. That's what I was talking about in that obscure sentence.
And unfortunately, just because there is an understanding of why and how, but the soul is asking for beauty))))
 
Anatolii Zainchkovskii:
Interesting, let's see. But it seems that the target for the first (left) trend is not defined correctly, the middle point should be on the previous bar.

No. The midpoint should be on the last bar touching the price - trust me ... In general, the price movement itself up to the target should ideally be accompanied by a trawl, which is activated after overcoming 80-85% of the predicted movement and has a value of 10-15% of the maximum. But it (the trawl) doesn't exist yet ... :):):) To be honest it's better without it. But it's only if you're constantly in the market.

I am writing a proper trading robot step by step, but some abstract conditions appear and they are difficult to implement in the code. As I adapt the methods I come up with some ideas how to avoid abstract stuff but it takes too long to write it. For example, the indicator on the screenshots (arrows and lines with the first impulse was written about a month ... :()

 
Anatolii Zainchkovskii:
As for time forecasts, they should go along with price forecasts. And if the price forecast works out of time then I consider it an accident, while a true forecast should confirm both price and time. That's what I was talking about in that obscure sentence.
And unfortunately, it's because we understand why and how, but our souls ask for beauty)))

Beauty is good! But it is different and perceived differently by different people.

Just wait a bit. That's the method number 2 will lay out - then the method of association will be able to get exactly what you need. But I'll tell you right away the amount of potential profit will be reduced by about 3 times, but about that later.

I have some work to do ... :)

 
RomFil:


In general, I am writing the robot in small steps, but as I write there are some abstract conditions that are difficult to implement in the code. As I adapt the methods, I come up with ideas on how to get away from the abstract - but it takes a long time to write it that way. For example, the indicator on the screenshots (arrows and lines with the first impulse was written about a month ... :()

In order to avoid abstraction we need to deduce a formula. In fact we have everything for that. There is a trend line that on every bar has a static increment calculated from the first impulse available. There is a number of the last bar which touches this trend line. Knowing the number of bars from the extremum to the trend line and the size of the increment of the trend line, a forecast price level can be calculated.
1234567
New comment