Incollection,
Explaining Supply and Demand: Stucture vs. Strategy
Abstract Book of the XXIII IUPAP International Conference on Statistical Physics, Genova, Italy, (9-13 July 2007)
Abstract
In the last half century economics has focused increasingly on the
strategic
interactions of agents, embodied implicitly or explicitly in terms
of game
theory. There are some situations, however, where other factors
may dominate,
such as the emergent dynamics of the interactions of the players,
or structural
constraints on their interactions such as those imposed by
institutions.
Standard examples of social systems where this is true are traffic
and crowd
dynamics; in both cases the useful theories are much more like
models of fluid
flow than neoclassical economics.\\
I will discuss some new examples involving standard problems in
economics. In
particular, I will present a theory for market impact in financial
markets,
which is a close cousin of the excess demand function (i.e. demand
- supply).
The theory is based on understanding how individual fluctuations in
supply and
demand aggregate in time. The underlying idea that explains the
shape of the
market impact function for large numbers of transactions is a
generalization of
the central limit theorem. Testing the theory against data from
the London and
New York stock exchanges shows that the data fit the model
extremely well.
Strategic interactions as reflected in market efficiency cause
important
adjustments in the predictions of the model, but the dominant
effect comes from
understanding the structure and dynamics of the aggregation process.
