Log in

Phase Space Singularities

Phase Space Singularities

Last updated 3146 days ago by JohnLast


The theory is that the market is not a signal. What we have is a multidimensional phase space. In those multidimensional space there are fractal attractors. There are many of them there are simple and complex (point attractors, cycle attractors etc.). For example when we see some number of cycles that does not mean that we have a signal with periodicity and frequency etc blah blah blah. The theory is that we have an underlying phase space wherein some attractor produce those cycles.

The fun thing is that the complexity of the phase Space is changing. Sometimes the phase space is relatively simple and sometimes it is really complex with its multidimensionality.

When the phase space is extremely simple we observe a Low Phase Space Singularity. When the phase space is extremely complex we see High Phase Space Singularity.

I call it singularity because the market dynamics start to be different from the normals market conditions. 

Low phase space singularity. What is typical of that is that many different algorithms are able to find the same solution simultaneously and act accordingly. Is it an accident that the Brain trend, the ASCTrend stops with digital smoothing, the ASCtrend signal, the Trend magic with or without digital smoothing and the SSAsqueeze find simultaneously the same solution?

You see brain trend had good results, ASCTtrend had good results, Trend Magic had good results. A simple moving average will have good results. The human trader judgement is blown out the statistics may be blown out and we can go against the trend and can be hurt. 

The idea is that when we have a relatively simple phase space of the possible solutions the algorithms are able to find simultaneously a solution. The algorithms are able to cooperate, but the humans we do not cooperate (one guy think it is oversold, the other it is overbought, one guy has long term horizon the other has short term).

And if FGDI (FDI) is in red (iVAR below 0.5) at both 15 and 30 m. time frame is a good approximation of the possible Low phase space singularity.

This Low phase space singularity is scary for the public policy makers, because all the market participants start to have the same horizon simultaneously. They are not able to adjust the economies to the market neither to guide the market as efficient as they would like.

They even may not have an idea what is going on and how their direct expensive interventions fail one after another on the Forex market.

Sometimes the Singularity goes into one direction, sometimes we have a series of ping - pong movements. All that is highly unpredictable several movements into the future, try to train a Neural Net to predict those market conditions and you will see, in fact whatever predicting algorithm you try to implement you will fail predicting the low phase space singularity. You have to react as quick and intelligent as possible, and even a SMA is good enough. 

The High Phase Space Singularity is exactly the inverse phenomenon. The Phase Space is terribly complex: nobody is right. The market gos up and down up and down. In that type of market conditions the statistical methods are the best. The Gaussian model approximates very well the market during those times.