Fractal structure analysis of NG, BR, GOLD, SILV, SPY, ED, ...

Collapse

Collapse

Does anyone interested in daily-base fractal structure analysis?

Фрактал - fSi [M3] [Word].png


At any given moment the market is in a large number of different-scale movements with their own regularities. Your practical actions significantly depend on your location in dynamic structures.

Explanation:


Example:


I'm thinking about creation of restricted Telegram channel (i.e. small private club of inquiring traders).

Currently I have only Moscow exchange's market data. So in addition to all those local instruments there are also some "international" futures on:

- Natural gas
- Brent oil index
- Gold
- Silver
- SPY
- EUR/USD​

What instruments (among this list) do you trade/prefer? And what approximate amount of $ per month you think you can donate for good personal analytics?

You are welcome to give me an answer in the comments or to contact me directly.
 
Last edited:
I would be interested in someone trying to explain why they think fractal patterns are an integral, unavoidable part of price movement.

There's a lot of craziness in technical analysis where someone picks a variety of mathematical functions, runs a fit of the parameters to make them fit a data set, and then goes "look I have a model for something". I see a lot less effort put into exploring why that particular model matters.

Are major traders using fractal model to make buy/sell decisions? Is there something inherent to supply or demand that is well modeled by fractals?
Simply put, why do you think fractals are worth paying attention to in the first place?

That could be an interesting discussion.
 
Collapse said:
Does anyone interested in daily-base fractal structure analysis?

View attachment 324637

At any given moment the market is in a large number of different-scale movements with their own regularities. Your practical actions significantly depend on your location in dynamic structures.

Explanation:


Example:


I'm thinking about creation of restricted Telegram channel (i.e. small private club of inquiring traders).

Currently I have only Moscow exchange's market data. So in addition to all those local instruments there are also some "international" futures on:

- Natural gas
- Brent oil index
- Gold
- Silver
- SPY
- EUR/USD​

What instruments (among this list) do you trade/prefer? And what approximate amount of $ per month you think you can donate for good personal analytic?

You are welcome to give me an answer in the comments or to contact me directly.
More...
Interesting. You should also cover at least all major 7 FX pairs or better all 28 majors and minors. Other interests than liquid FX I do not have at this moment. But SPY and Gold, Oil, also T-note is also interesting. Where can I see what you are publishing ? I deleted my old telegram. I would give it a try. Looks interesting to me. But I need to check and test first. Do you offer these curves as indicator for Metatrader5 too ? Any mql seller link you have on its store ?
 
There is a good study on this.

https://www.sciencedirect.com/science/article/abs/pii/S0378437115005506

Developing trading strategies based on fractal finance: An application of MF-DFA in the context of Islamic equities

Abstract
We provide a new contribution to trading strategies by using multi-fractal de-trended fluctuation analysis (MF-DFA), imported from econophysics, to complement various momentum strategies. The method provides a single measure that can capture both persistency and anti-persistency in stock prices, accounting for multifractality. This study uses a sample of Islamic stocks listed in the U.S. Dow Jones Islamic market for a sample period covering 16 years starting in 1996. The findings show that the MF-DFA strategy produces monthly excess returns of 6.12%, outperforming other various momentum strategies. Even though the risk of the MF-DFA strategy may be relatively higher, it can still produce a Sharpe ratio of 0.164, which is substantially higher than that of the other strategies. When we control for the MF-DFA factor with the other factors, its pure factor return is still able to yield a monthly excess return of 1.35%. Finally, we combine the momentum and MF-DFA strategies, with the proportions of 90/10, 80/20, and 70/30 and by doing so we demonstrate that the MF-DFA measure can boost the total monthly excess returns as well as Sharpe ratio. The value added is non-linear which implies that the additional returns are associated with lower incremental risk.
 
We have discussed fractals before. The main problem with fractals in trading markets is that "anything" can be a fractal. If you don't like those divisions on the chart that you drew in green you could always zoom in and out and go to a different time frame to find patterns that would satisfy your initial premise.

Then, you can explain past behaviors of a market and line them up with news. You would get to a point where no one could challenge you. You would become a master at explaining the past. But you would never be able to predict the future.

Pretty much like any technical analysis theory. You can extend yourself explaining the past with multiple calculations but that will never be a reason to explain the future.

If you look at any paper on fractals related to trading markets you will see that they are all free, written by Ph.D. students. The main reason is that they look for a complex subject that lets them apply everything they learned during the course. It is a perfect subject to fill thousands of pages, essentially because no one can challenge you. If you choose the right timeframe you will always find enough examples to sustain your theories.

Everyone can be a master when they know what is going to happen. That is essentially what every book about technical analysis is based on. They are full of examples of how to trade signals based on past behaviors, they carefully choose the perfect chart to prove their points. They never show charts when their point is denied, that would be an actual good book.

Fractals make this 'confirmation bias' problem get to an extreme level. The mathematics involved with fractals and chaos theories make them very attractive to the eye, thinking that they are actually on a good path to predict an outcome, but they are always just explaining the past.

Particularly when using intra-day data, there is so much noise that anything can be understood as a fractal and fit perfectly in your initial premise.

IMO it is interesting to look at, but never to create signals to trade. If you are successful there will be a great weight on luck there.
 
engineering said:
I would be interested in someone trying to explain why they think fractal patterns are an integral, unavoidable part of price movement.

There's a lot of craziness in technical analysis where someone picks a variety of mathematical functions, runs a fit of the parameters to make them fit a data set, and then goes "look I have a model for something". I see a lot less effort put into exploring why that particular model matters.

Are major traders using fractal model to make buy/sell decisions? Is there something inherent to supply or demand that is well modeled by fractals?
Simply put, why do you think fractals are worth paying attention to in the first place?

That could be an interesting discussion.
More...

While everyone is dealing with the price, I'm working with its boundaries. That's why I call this approach "unique".

Did you see my real-time example? It worked. Do you have any idea "why"?
 
TrAndy2022 said:
There is a good study on this.

https://www.sciencedirect.com/science/article/abs/pii/S0378437115005506
More...

Okay, I have not read the research paper. I briefly skimmed the abstract. I am not especially interested in fractal theory as it applies to trading.

But why in the world would they use "a sample of Islamic stocks listed in the U.S. Dow Jones Islamic market"??

Are they referring to the Dow Jones Islamic Market World Index?

I didn't know that such an index existed until I saw this paper.

The index has 4810 constituents, and the top ten by weight are all in the S&P 500.

The fact sheet says that "this index measures the performance of stocks traded globally that pass rules-based screens for adherance to Shari'ah investment guidelines."

And that's kind of interesting. I did not know that there were Shari'ah investment guidelines.

But if they actually want people to read the research paper, they probably should have just said they were using a sample of stocks in the S&P 500. An awful lot of people are simply not going to get past the title of this paper.

And the choice of words matters. What is an "Islamic stock?" And what is the "U. S. Dow Jones Islamic market"? If they are referring to the index, then they should use the terminology that is used by the index.
 

Attachments

Collapse said:
While everyone is dealing with the price, I'm working with its boundaries. That's why I call this approach "unique".

Did you see my real-time example? It worked. Do you have any idea "why"?
More...
Please post more examples. Actual ones please, also when it is not so good working to get an honest picture of it. Because there are always drawdowns in every strategy. One needs to be aware of them.
 
To be fair, most research papers are written by Ph.D students.

I have been trying to understand dynamics/dynamic systems, I have the book Nonlinear Dynamics and Chaos: With Applications to Physics, Biology, Chemistry, and Engineering. It is really different and hard to understand though. There is the whole field then of Economic Dynamics but I can't even get the basics down to get to that.

The main problem I would say is that hand drawing on charts just has nothing to do with the research that has been done in this area. That drawing is not a fractal but highlighting hierarchical structure. That could be an interesting area to look into but it is not a fractal.

Personally, I think the scaling properties of the markets are anything but fractal. Bifurcation from dynamics,
a sudden change in the qualitative behavior of a system, is what I think is probably the more applicable topic to markets from dynamics. Attractors/Repellors probably make sense too but all this is hard to understand.

Drawdown Addict said:
If you look at any paper on fractals related to trading markets you will see that they are all free, written by Ph.D. students. The main reason is that they look for a complex subject that lets them apply everything they learned during the course. It is a perfect subject to fill thousands of pages, essentially because no one can challenge you. If you choose the right timeframe you will always find enough examples to sustain your theories.
More...
 
You must log in or register to reply here.
Back
Top