Topological analysis

utilizing the complexity and chaos theory concepts in trading system development

Moderator: moderators

Paul
rank: 150+ posts
rank: 150+ posts
Posts: 222
Joined: Tue Aug 19, 2008 1:00 pm
Reputation: 0
Gender: Male

Topological analysis

Postby Paul » Tue Aug 19, 2008 1:59 pm

Hi there,

I analyse FX through fractal modelling which I have been using for more than ten years. It is different than Bill Williams commercial software. He either has not found it or does not want to disclose it. The main difference is in the making of fractals. One would expect to be able to determine a fractal structure for the future, not just ex post. To establish future turning points one needs a complete fractal image, not a fragment which belongs to the past.

There is a unique and obvious (to me) mathematical link between the Fibonacci numbers/ratios, the Elliott fractals and the Feigenbaum's universal value (4.669...). I am talking about a link in 2-d space, of course. The theory of chaos is around.

For a couple of months I have tested the interbank volume (or rather ticks as explained by Reuters 3000) on FX for signals of turbulence similar to turbulences found in the physics of fluids and gases. A streamline of quotes can be treated like a flow, though discreet in nature. This dynamic process should not be free from erratic behaviour and may develop local turbulences in the flow. A turbulence is a highly unstable state of the system, and the only way for ithe system to return to stability is through dramatic changes of prices (reversals). It follows that if turbulences occur then contrarian positions should bring a handsome profit. It does not matter what causes a turbulence. Even physisists argue about that. What is important is the fact that the flow of quotes sometimes slows down or accellerates to such an extent that creates a turbulent flow.
My question is whether someone has been studying the interbank volume from this angle. I currently concentrate on just two pairs EURUSD and USDJPY for lack of time and analyse M30 volumes. During a session I spot a few turbulences for each pair. USDJPY is my best indicator for the behaviour of the market immediately after the US data, excellent for forecasting EURUSD.
As for today USDJPY signalled BUY at 9:00 and since no other signal occurred it was valid for today's PPI and home market report. Before 14:30 I went long on USDJPY and short on EURUSD and closed 30 minutes later.
In the attachments you will kindly find all turbulences spotted by me for the given period marked by red arrows. The direction of an arrow is determined not guessed. Each arrow is born at the close of an M30 candle.
You may kindly notice that the market reaction is not always immediate.
Attachments
2.gif
USDJPY M30 chart with signals of turbulence.
2.gif (21.65 KiB) Viewed 1265 times
3.gif
EURUSD M30 chart with signals of
turbulence.
3.gif (22.67 KiB) Viewed 1265 times

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.

Paul
rank: 150+ posts
rank: 150+ posts
Posts: 222
Joined: Tue Aug 19, 2008 1:00 pm
Reputation: 0
Gender: Male

Topological analysis

Postby Paul » Wed Aug 20, 2008 11:36 am

Hi there,

I have updated signals of turbulence.
EURUSD signalled SELL, shot up almost 40 pips and returned to the signal level. I added positions at the top.

USDJPY did not perform in quite the same manner like EURUSD. I added positions at the lows.

I have closed all positions now with a good profit on balance.
Attachments
4.gif
4.gif (22.27 KiB) Viewed 1239 times
3.gif
3.gif (22.74 KiB) Viewed 1239 times
4.gif
4.gif (22.26 KiB) Viewed 1243 times
3.gif
3.gif (22.79 KiB) Viewed 1254 times
4.gif
4.gif (22.24 KiB) Viewed 1254 times

Paul
rank: 150+ posts
rank: 150+ posts
Posts: 222
Joined: Tue Aug 19, 2008 1:00 pm
Reputation: 0
Gender: Male

Topological analysis

Postby Paul » Wed Aug 20, 2008 3:25 pm

Hi there,

I have given you a live example of what the market is doing when a turbulence occurs.

There is an immediate or a delayed reversal of prices.
There is a need to open a number of positions, not just one or two. There is a strategy for that. Stop loss is remote, so the risk/reward ratio may be high.

As you see from charts EURUSD has outperformed USDJPY, but in both the cases profit was handsome.
Attachments
3.gif
3.gif (22.5 KiB) Viewed 1236 times
4.gif
4.gif (22.38 KiB) Viewed 1236 times

heliosphan
rank: <50 posts
rank: <50 posts
Posts: 2
Joined: Thu Aug 28, 2008 2:30 am
Reputation: 0
Gender: None specified

Postby heliosphan » Thu Aug 28, 2008 3:10 am

Hi Paul. Could you share the indicators or save your setup as a template? I'd like to learn more about this method and try it out

I have done some casual research in the area of quantum mechanical systems. What you describe in some ways sounds similar to a program called J-chart http://www.j-chart.com/

The problem with it, not easy to interpret and the documentation is crappy. you can get a free demo. It's based on what the author calls equilibrium and chaos, similar to what you call turbulence versus non-turbulent pricing. It refers to the quantum mechanic theory called pair production for the basis of how its strategy works

http://physics.pdx.edu/~egertonr/ph311-12/pair-p&a.htm

a pair of elementary particles (a particle and its antiparticle of the same mass but opposite electrostatic charge) are created from the energy (hf) of the original photon. In this case, the two particles are an electron and an antielectron (more commonly known as a positron, whose rest mass m0 is the same as that of an electron but whose charge is +e ).


based on this description you can easily compare it to how the markets move. particle/antipartcle is either side of a potential trade, while
the energy of the original photon is our universe, i.e. currency pair



Pair production can be represented by an equation which represents the conservation of total energy (or mass-energy):

hf = 2(m0 c^2) + K(-e) + K(+e)

Here, (m0 c^2) = 0.511 MeV is the rest energy of an electron, which is equal to that of the positron, so the factor of 2 represents the fact that two particles of identical rest mass are created. K(-e) and K(+e) represent the kinetic energy of the electron and positron, immediately after their creation.



If the photon energy were exactly 2m0 c^2 = 1.02 MeV, the two particles would be created at rest (with zero kinetic energy) and this would be an example of the complete conversion of energy into mass. For photon energies below 2m0c^2, the process cannot occur; in other words, 1.02 MeV is the threshold energy for pair production. For photon energies above the threshold, a photon has more than enough energy to create a particle pair and the surplus energy appears as kinetic energy of the two particles.



There is, however, a further condition which must be satisfied during the pair-production process: conservation of momentum. Taking this requirement into account, we can anticipate that pair production cannot take place in empty space; something must absorb the momentum (p=h/l =hf/c) of the initial photon. (To see this, consider the threshold situation where the particles have to be created at rest and cannot themselves absorb any momentum). The photon momentum can be absorbed by an atomic nucleus, which is thousands of times more massive than an electron or positron and can therefore absorb momentum without absorbing much energy; therefore the energy-conservation equation above remains approximately valid. Consequently, pair production is observed when high-energy gamma rays enter a solid, where a high density of atomic nuclei is present.



There exists an inverse process to pair production called pair annihilation, in which a particle and its antiparticle collide and annihilate each other, the total energy of the two particles appearing as electromagnetic radiation. In the case of an electron and positron, the energy balance can be represented as:

2m0c^2 + K(-e) + K(+e) = 2 hf

The first term represents the rest energy of both particles, the second and third terms are the kinetic energies just before the collision, and the term on the right-hand side of the equation represents the creating of two photons, each having a the same frequency f and energy hf . The equation indicates that hf must be at least equal to m0c^2=0.51 MeV, so the photons produced correspond to g -ray radiation. Like pair production, this process usually takes place inside a solid, which provides the electrons which are needed. The positrons enter the solid as a beam of particles produced by a particle accelerator.

heliosphan
rank: <50 posts
rank: <50 posts
Posts: 2
Joined: Thu Aug 28, 2008 2:30 am
Reputation: 0
Gender: None specified

Postby heliosphan » Thu Aug 28, 2008 3:19 am

More quotes from site:
If the kinetic energies of the two initial particles were both small (<< m0c^2), the total momentum before the collision would be close to zero. From conservation of momentum, the momentum after annihilation must also be approximately zero, and the only way that this can happen is for the two photons to be emitted in opposite directions such that their individual momenta cancel. More typically, the positrons entering the solid have appreciable momentum whereas the electrons (originally present in the solid) are almost at rest. Then the paths of the two g -ray photons are at some angle to each other, and these paths intersect at the position of the pair-annihilation event (see Fig. 4-xx).




This is the principle behind positron emission tomography (PET), a medical diagnostic technique which provides images showing the internal structure of body organs. By detecting the direction of travel of g -ray photons which are produced simultaneously (coincidence detection) and counting the number produced in different regions of space, the PET machine builds up an image of the electron density within a plane (actually a narrow slice) within the tissue. Moving the detectors (or the patient) forward, the process can be repeated so that three-dimensional information (a tomograph) is obtained. The positrons are provided by a radioisotope, injected into the bloodstream. PET is now a commercially developed technique, complementary to CT and MRI scanning.

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.

xtremeforex
rank: 50+ posts
rank: 50+ posts
Posts: 114
Joined: Mon Nov 16, 2009 7:43 pm
Reputation: 0
Location: Toronto
Gender: None specified

Postby xtremeforex » Wed Nov 17, 2010 4:37 pm

Does anybody know if MATLAB has forex functions?

I would love to test Paul's theory as well. I do understand the ideas of fixing turbulence using turbulence in terms of super-position and cancellation effects of waves.

User avatar
TheRumpledOne
rank: 10000+ posts
rank: 10000+ posts
Posts: 11416
Joined: Sun May 14, 2006 9:31 pm
Reputation: 45
Location: Oregon
Real name: Avery T. Horton, Jr.
Gender: None specified
Contact:

Postby TheRumpledOne » Thu Nov 18, 2010 7:37 pm

This must be where the Yale students hang out...LOL!
IT'S NOT WHAT YOU TRADE, IT'S HOW YOU TRADE IT!

Please do NOT PM me with trading or coding questions, post them in a thread.

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Re: Topological analysis

Postby Relativity » Fri Nov 19, 2010 4:29 pm

Paul wrote:Hi there,

I analyse FX through fractal modelling which I have been using for more than ten years. It is different than Bill Williams commercial software. He either has not found it or does not want to disclose it. The main difference is in the making of fractals. One would expect to be able to determine a fractal structure for the future, not just ex post. To establish future turning points one needs a complete fractal image, not a fragment which belongs to the past.

There is a unique and obvious (to me) mathematical link between the Fibonacci numbers/ratios, the Elliott fractals and the Feigenbaum's universal value (4.669...). I am talking about a link in 2-d space, of course. The theory of chaos is around.

For a couple of months I have tested the interbank volume (or rather ticks as explained by Reuters 3000) on FX for signals of turbulence similar to turbulences found in the physics of fluids and gases. A streamline of quotes can be treated like a flow, though discreet in nature. This dynamic process should not be free from erratic behaviour and may develop local turbulences in the flow. A turbulence is a highly unstable state of the system, and the only way for ithe system to return to stability is through dramatic changes of prices (reversals). It follows that if turbulences occur then contrarian positions should bring a handsome profit. It does not matter what causes a turbulence. Even physisists argue about that. What is important is the fact that the flow of quotes sometimes slows down or accellerates to such an extent that creates a turbulent flow.
My question is whether someone has been studying the interbank volume from this angle. I currently concentrate on just two pairs EURUSD and USDJPY for lack of time and analyse M30 volumes. During a session I spot a few turbulences for each pair. USDJPY is my best indicator for the behaviour of the market immediately after the US data, excellent for forecasting EURUSD.
As for today USDJPY signalled BUY at 9:00 and since no other signal occurred it was valid for today's PPI and home market report. Before 14:30 I went long on USDJPY and short on EURUSD and closed 30 minutes later.
In the attachments you will kindly find all turbulences spotted by me for the given period marked by red arrows. The direction of an arrow is determined not guessed. Each arrow is born at the close of an M30 candle.
You may kindly notice that the market reaction is not always immediate.


Paul,

I appreciate such a research, but can you simplify it? What does all this eventually conclude? It does not make common sense without a simple conclusion.

User avatar
ajaymein
rank: 500+ posts
rank: 500+ posts
Posts: 885
Joined: Sun Nov 09, 2008 6:34 am
Reputation: 0
Gender: None specified

Postby ajaymein » Fri Nov 19, 2010 4:37 pm

This was over 2 years ago...somehow I think he may of abandoned this...thread at least ?..

xtremeforex
rank: 50+ posts
rank: 50+ posts
Posts: 114
Joined: Mon Nov 16, 2009 7:43 pm
Reputation: 0
Location: Toronto
Gender: None specified

Postby xtremeforex » Fri Nov 19, 2010 4:59 pm

TRO,

You make me laugh lol.

I know this stuff is overcomplicating trading but it is kinda fun to see if patterns can be found through numerical analysis. It's kind of the "you never know" what's hiding inside the data.

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.


Return to “complexity & chaos theory”