Next generation behavioral finance
Underlying next-gen behavioral finances are theories such as Herding and Network effects
General agents interacting and competing in financial markets inherently results in:
- positive feedback
- faster than exponential price dynamics (ultimately unsustainable)
- identifiable phases of instability
- Extensive research indicates that the nature of price patterns inherently reflects the nature of humans who are collectively trading assets with uncertain value
- Agent-based models with generic rules produce the same structures and can be used to reverse-engineer financial markets. They also provide a framework for testing hypotheses and feasibility
- Computational engines at SIMAG® build on the rational integration of our understanding of how interacting and competing agents with feedback dynamics interpret and act on uncertain information
Professor Didier Sornette
Scientific foundation & examples
What do a high-pressure-tank on a rocket, a seismic fault, and a busy market have in common? All three have the potential for extreme behavior: rupture, quake, or crash.
The research carried out by Prof. Didier Sornette shows they can all be described in very similar basic physical terms. They can be described as self-organizing systems that develop similar patterns over many scales, from the very small to the very large. The same mathematical framework of “finite-time singularities”, provides a very powerful language and description of these phenomena. Prof. Didier Sornette and his group have become leading experts on this mathematical framework, which provides a unifying language for systems undergoing changes of regimes, transitions, transformations, revolutions, crises, ruptures, etc.
The over-arching philosophy of the theory powering SIMAG® methods is the recognition that the world, especially social structures and financial markets, are organized as complex hierarchical systems. Multiple systems interact and influence each other, leading to recognizable (finite-time singular) behaviors associated with pockets of predictability.
Example 1: Rocket science (literally!)
Damage and fracture of materials are major areas of interest because of their economic and human costs. These topics cover a wide range of phenomena such as the cracking of glass, the aging of concrete, the failure of fiber networks, and the breaking of a metal bar subject to an external load. All of these phenomena involve composite systems of various components. Failures of composite systems are of utmost importance in the naval, aeronautics, and space industries.
Thus,when Prof. Didier Sornette invented a method to predict the failure of pressure tanks strapped on to the Ariane rocket and satellites, it quickly became routine in industrial monitoring procedures. The key concept of this invention is to quantify the collective behavior of acoustic emissions that reveal the level of degradation and the potential for a systemic failure within the pressure tank. Mathematically, the method uses concepts from the “renormalization group theory” of critical phenomena, for which a Nobel prize in physics was awarded to Kenneth Wilson in 1982.
Example 2: Social Systems
Social systems, in particular financial markets, are organized and structured by the interplay of exogenous information flow, perturbations, and shocks on the one hand, and by endogenous interactions that react to external solicitations on the other hand. Book sales, YouTube video views, movie attendance, and social media hype are illustrative examples in which the ETH Zurich group associated with SIMAG® has demonstrated the existence of a small number of universal classes in the response function to shocks, news, and disruptions.
These classes can be used to distinguish endogenous from exogenous anomalies. This in turn enables the quantification of the response of the system, its predictability, and by extension the operational implementation of optimal strategies.
More examples of self-organizing systems
- financial bubbles and crashes
- formation of lightning
- planet formation in the solar system by run-away accretion of planetesimals
- singularity theorems of Penrose and Hawking in General Relativity
- formation of black holes (general relativity coupled to a mass field)
- plasma physics within tokamaks for nuclear fusion
- genesis of weather fronts in meteorology
- description of hydrodynamic turbulence
- rupture and failure of materials and engineering structures
- earthquake nucleation processes
- chemotaxis of micro-organisms aggregating to form fruiting bodies
- surface instabilities to form spikes
- snap-off of fluid droplets in the kitchen
- Euler’s disk (rotating settling coin)
Log periodic power law singularity (LPPLS)
Resulting analytics identify the degree of system (in-)stability – a highly non-stationary indicator providing probabilistic information on the attractiveness of owning any given stock, sector, or country at any given time
Proprietary research identifies complex “fingerprints” of regime shifts that have log periodic power law singular form (LPPLS)
Parameter ranges, statistical confidence metrics, and dynamic patterns are used to diagnose the status of the overall process, the system (in)stability, and the level of risk
Resulting metrics deliver robust indicators of longer term market regime (risk on, risk off). These metrics are obtained at the cost of intensive computational investment
The research of Professor Didier Sornette has shown that in finance, most bubbles, crashes, and regime shifts have an internal cause or origin (i.e. they are endogenous)
Present generation behavioral finance focuses mainly on the biases of decision making in individuals, including over-confidence, the conjunction fallacy (i.e. the assumption that specific conditions are more probable than a single general one), the disjunction effect (i.e. the breaking of the sure thing principle), the Allais paradox (i.e. the inconsistency of actual observed choices with the predictions of expected utility theory), and the Ellsberg paradox (people overwhelmingly prefer taking on risk in situations where they know specific odds rather than an alternative risk scenario in which the odds are ambiguous).
Our approach is “next-generation behavioral finance”: We identify and use the biases and structures resulting from the collective behavior of humans and their interactions. This leads to phenomena such as “the wisdom of crowds” as well as “the madness of crowds”. As it concerns marketable securities and financial markets, next-generation behavioral finance focuses on collective dynamics which are are the aggregate of many individual decisions, and thus more relevant.