Explain about Fibonacci... - page 3

 
Vinsent_Vega >>: - an impression that Fibo levels don't work most often appears because they are always triggered with "outliers"... If you're collecting statistics on levels, you should consider, say, not a hard 38.2 level - but 38.2 +/- X p.

There are many opponents of Fibo. They say that any price extremum on the history can be compared to the Fibo level of some swing. And on this de facto basis, even if you point to the sky, there is still the required Fibo level - on the history.

Well, it is so in principle. In order to predict a future extremum, the single TF is not enough. It is necessary to look not only at the current one, but at a smaller and at a larger one. Each of them has its own ZZ and respective swings. But the number of swings that affect the situation at a certain moment in time is always limited (on average no more than 5-7 on each TF, but sometimes it can be larger, up to 10).

Now, if you plot all the Fibo levels from each influencing swing and on each TF (current+lower+older), you get a pretty dense grid of levels, for which it's almost like this: if you point your finger anywhere on this grid, there will still be some Fibo level from some swing nearby. The levels themselves, by my estimation, turn out to be several hundred.

How should we orientate ourselves in this grid? We should look for the levels, near which as many Fibo levels as possible are located. Such areas are called Fibo-clusters. These are potential levels of future price extremums. In order to select the most probable one among several, we will have to assign different weights to the levels from the corresponding swings so that the cluster weights are different. How to do it in a more or less reasonable manner - I do not know.

Technically this "indicator" is very complicated, and there are a lot of pitfalls and hidden parameters, but I believe some programming genius could write it. And I also assume that it could be used for profitable trading. But I haven't heard of any realised attempts, apart from Fibo-cot from Alpari. It has not only levels but also Fibo fans. Of course, I should also mention nen with Gartley's butterflies, but as far as I know the system is not fully automated in nen as well.

Wave theory, in my opinion, is not needed at all in such a system. I haven't read Fisher, but I have read Meyer.

 
There is little statistical evidence to support the 'workings' of the Golden Ratio. To get more convincing data it is necessary to complicate the conditions for collecting statistics. The statistics need to be collected dynamically, in a tester. The reference to ONICS above partially describes conditions of such additional experiment that would be desirable.
 

Fibonacci levels are probably the subtle difference between financial time series and martingale (and incidentally Wiener process).

 
There is also statistical evidence to support the frequent occurrence of ab=cd structures. This structure stands out clearly. But applying it "head-on" is probably not possible.
 
sayfuji писал(а) >>

Gunpowder was known in China as early as the beginning of our era. And there was no mathematical calculation of the number of reagents. The Chinese had no idea about the composition of potassium nitrate, but actively used it and burned it to death.

Good example, but off-topic. That's right:

Let a certain people in a certain country persistently set fire to, for example, concrete for a millennium, just because their religion says it is the best thing to burn! Even though their science has shown for a hundred years that there is nothing to burn in concrete, and as a consequence no gunpowder from it!

So, all this fuss with F levels and E waves is obviously closer to concrete gunpowder and total religion than to the real invention of gunpowder and its operation without knowledge of the laws of chemistry at the modern level. I have at one time typed statistics on the significance of these levels - no result, or rather the same as for random Brownian one-dimensional motion.

Vinsent_Vega wrote >>


arguing with Neutron is like going against the wind...

Thank you. I made a fucking comparison...

 

The eurusd has built a lot of zigzags on the minute history... have chosen pentovolves. A complete movement is a five-wave movement. So the first wave has a clear maximum of 62% of the total movement in most cases. I will not give you charts and instrucƟons. I did not get a permission for that.

===================

Also, such pictures do not allow us to say that the fibs (golden ratio) do not work.

 

That's right:

In all countries of the world, people, making assumptions about the composition of the objects around them, find a pattern in the fact that many objects, including non-manufactured objects, obey a certain law, the law of equilibrium and harmony. It was proved that gunpowder and cement have similar constituent elements, e.g. carbon. Cement had not been ignited by anyone and it had not been possible to prove its impossibility either. On the other hand, gunpowder burns well in spite of (or due to) the presence of the same carbon. Final conclusions about the structure and the nature of the constituent elements of the environment and their origin have not been drawn, but the existing regularities allow us to obtain some results. This is the first step of a long journey.

 

Many thanks to everyone who responded! But that's the beauty of it - I "played around" with the story, and the coincidences I saw were astonishing to me, because I'm a realist by nature and I need to "feel" everything to believe it. In addition I was skeptical that one can draw Fibonacci lines (for me they were just lines at first) and see the regularities according to which the market moves. Well, after successful experimentation with history I have only one question - what will happen next? Everything turns out very beautifully on history and I would say it is logical (of course, you should not take these words literally), but how can we apply Fi to present and future? After all, if we apply only to history, the numerical pattern of Fibonacci in the market means absolutely nothing.

 
YuraZ писал(а) >>

I like very much e.g. 161 level

You put a Fibo level on a worked zigzag and then you usually reach the 161 level in as many cases

statistically

---

it's almost a pattern.

Do you mean the ZigZack present in MT4 tools (indicators)?

Does it matter if the ZigZag is ascending or descending?

What if it is underdrawn? In other words, it was drawn downwards at first, and then the trend began to move downwards and, consequently, it will continue drawing the ZigZag, is it normal?

 
Mathemat, do you use Fibo in trading?