CHAOS MONITOR

ChaosMonitor™ is a pioneering investment research service focusing on the nonlinear endogenous risk factors of the financial markets

• 'Efficient market hypothesis' has been proven wrong and needs to be replaced.
• Financial markets should be modelled as a complex dynamical system. Modern physics has developed a large body of knowledge about such systems that can be directly applied to market forecasting. Complexity Theory, Chaos Theory and Dynamical Systems Theory are names of the related branches of applied mathematics.
• As a complex nonlinear system, the market is essentially deterministic, not random, even though the system's complexity makes it impossible to predict the future state most of the time (and during this time a Gaussian model provides a passable approximation).
• It is possible to predict short-term behaviour of a complex dynamical system by studying its nonlinear endogenous properties (self-similarity, log-periodic power-law scaling etc).
• One source of predictability is 'self-organised criticality', i.e. the system's propensity to arrive at a critical state just before a major change in its behaviour.
• Another source of predictability is the fact that the trajectory of a complex dynamical system can settle in a stable region (so called attractor) for a period of time.
• Studying endogenous risk factors as a complex nonlinear system is another form of fundamental analysis, not unlike the analysis of global macro imbalances or company balance sheet risk.

'Bubbles' are emergent market phenomena that appear 'by themselves' and not due to some external shock, such as change in fundamentals or expectations. A random small price rise can suddenly cause massive execution of limit orders, option hedging and bandwagon activity of trend followers -- all of which would force the price even higher. At some critical point the rapid rise will end and the market behaviour will change sharply.

Today we can detect the 'signature' of an approaching market critical point that marks a trend reversal.

It has been established that extreme trend persistence creates unstable critical points in financial markets, characterised by precursory power law advances or declines in price, coupled with converging log-periodic cycles.

These critical points mark the end of a trend at a given time-scale (hourly, daily etc.) and are often followed by significant price reversals of matching scale.

For years patterns in market prices were called an illusion. Modern science views the markets as a complex nonlinear system. The trajectory of such systems can sometimes settle in a stable area called an attractor. Unlike so called 'Technical analysis' (which is an exercise in pattern recognition without access to a modern theoretical framework and mathematical apparatus), ChaosMonitor™ relies on advanced scientific algorithms and a lot of computer power to detect these pockets of market predictability.

A 'Price Channel' is a popular chartist tool, that is normally drawn by hand. The ChaosMonitor™ algorithm automatically detects system attractors as they appear and plots them in the 'channel' format that is familiar to market practitioners.	In Chaos Theory 'attractors' are temporary stable areas of the system's trajectory. The 'cobweb' diagram of a 'price channel' shown above (ΔxΔt vs x), illustrates price-time dependency that is impossible according to the 'efficient market hypothesis', but common for chaotic systems.

ChaosMonitor™ detects and presents tens of trading opportunities each day on intraday and daily time horizons. It accurately identifies points where market behaviour is about to change from mean-reverting to strongly trending (blue marks below), and where the current trend is about to stop (magenta signals). Market coverage includes major futures on interest rates, government bonds, equity indices and commodities as well as individual benchmark stocks (S&P100, EURO STOXX 50 and Hong Kong H-shares) and currencies in Americas, Europe and Asia. It is also possible to run our algorithms on your in-house data in real time. Please see our demo screencast for more examples, consult the frequently asked questions section, and request a free trial.

ChaosMonitor™ occupies a unique niche among investment research firms by focusing on market endogenous risk factors (source of returns due to nonlinear complex nature of the markets, not due to change in fundamentals etc.). Because of its unique methodology rooted in nonlinear physics, the model is a very effective Global tactical asset allocation (GTAA) and market timing tool. For example, our model has identified moments of extreme market instability weeks before market sell-off in 2008 (blue signals below), and showed when the market decline was reaching its extreme (magenta signal). Please see our demo screencast for more examples and consult the frequently asked questions section.

ChaosMonitor™ is a web application available through a regular browser (desktop or mobile device). We monitor over 2000 datasets of market instruments from major exchanges, process them in our datacentre and display resulting signals in real-time. Intraday, daily and weekly timescales are covered. Subscribers can set a filter for market sectors, timescales and regions they focus on. The service is delivered on a monthly subscription basis.

Per Bak, Chao Tang, and Kurt Wiesenfeld, "Self-organized criticality: An explanation of the 1/f noise," Physical Review Letters, vol. 59, no. 4, pages 381-384, 1987.
Hurst H.E. Long- term storage capacity of reservoirs, Trans. Of the American Society of Civil Eng. Nr. 1161951, P.770-799
Kaufman, S.K. A New Method of Forecasting Trend Change Dates. CYCLES, Sept/Oct 1990.
Lorenz, Edward N. Deterministic non-periodic flow, Journal of the Atmospheric Sciences, vol. 20, pages 130-141 (1963).
Mandelbrot, Benoit. The variation of certain speculative prices. Journal of Business, vol. 36, pages 394-419. 1963.
Mandelbrot, Benoit. How Long Is the Coast of Britain? Statistical Self-Similarity and Fractional Dimension, Science, Vol. 156, No. 3775, pages 636-638. 1967.
Mandelbrot, Benoit and Hudson, R.L. The (Mis)behavior of Markets: A Fractal View of Risk, Ruin, and Reward. 2004
Poincare, Jules Henri. Sur le probleme des trois corps et les equations de la dynamique. Divergence des series de M. Lindstedt. Acta Mathematica, vol. 13, p 1-270. 1890
Sornette, Didier and Johansen, A., Significance of log-periodic precursors to financial crashes, Quantitative Finance1(4), 452-471, 2001
Sornette,Didier, Discrete Scale Invariance and complex dimensions, Phsyics Report, 1998
Sornette, Didier, Critical Phenomena in Natural Sciences, Springer-Verlag, 2006

ChaosMonitor™ is developed and maintained by Rational Decisions, a research and software development boutique in the UK specialising in advanced decision algorithms for the investment industry. People behind the service are:

Igor Drozdov

Before founding Rational Decisions Igor was managing Global Tactical Asset Allocation portfolio at Mn Services, one of the largest Dutch pension fund managers. Earlier he was doing proprietary trading and market-making in emerging markets short-term interest rates products at JP Morgan, London. Igor began his career in 1994 as Latin American sovereign debt trader at ABN AMRO in Amsterdam and later became Head of fixed income trading and sales at one of the bank's subsidiaries.

S Kris Kaufman

Kris is the author of pioneering research on applying chaos theory principles to the markets. His has built a successful career as a geophysicist with a focus on 3-d oil exploration, seismic waveform modelling, and earthquake prediction before founding Parallax Financial Research, Inc. in the 1990. For over a decade Parallax has been providing cutting edge market analysis algorithms to a select group of prop-desks and fund managers.