The trading desks of major investment banks and hedge funds have long sought maths and physics talent to try to turn science into profit. And since the dawn of markets, alchemists of all stripes have been trying to build the ultimate device that will allow a few to profit consistently.
But in the 1990s, physicists began looking at ever more robust data coming from economics and finance as a way to look at market dynamics and other economic principles in a new light. The movement, for lack of a better term, was coined econophysics, although in recent years its popularity has waned.
Didier Sornette, a trained statistical physicist and geophysicist and now professor of finance on the chair of entrepreneurial risks at the Swiss Federal Institute of Technology in Zurich, does not seem bothered by the waning enthusiasm for the multi-disciplinary approach. Rather, he is doing what he has done for much of his career, which is publishing not only in leading physics journals but also leading finance ones.
Prof Sornette, author of Why Stock Markets Crash (2004), is essentially trying to better understand the dynamics of bubbles. He is alone - or one of very few - in leading three concurrent fields in the analysis of complex systems: pure physics, applied economics and econometrics, and market practitioners.
…
However, not all regime changes are crashes, it is important to note. Prof Sornette distinguishes between endogenous and exogenous crashes: the former account for two thirds of the cases he has studied, and show clear signs of bubbles characterised by superexponential growth that finish in a crash. The other third, exogenous crashes, show no signs of bubble activity or build-up phase.
It is this last third that Prof Sornette wryly claims obey the laws of economic theory, which wants to claim that crashes should result from some exceptional new piece of information being introduced. But for the two thirds majority - the endogenous ones - the crash does not need a new piece of information.
“This is where physicists have perhaps been at the forefront,” says Prof Sornette. “The understanding of the interaction between many atoms and molecules, which in physics we refer to as the ‘N-body problem’, requires information throughout the whole sample. It’s a phenomenon whereby the global collective system behaves as one - this is the hardest part for economists to grapple with.”