|
It’s a long way from the heady floors of the stock exchanges of the world, but in a small office overlooking the shores of Lake Burley Griffin at ANU, physicists are generating information the bankers in the bearpits would probably love to know.
 A small research team in a hybrid field known as econophysics, including Dr Tiziana Di Matteo (right), are applying methods from statistical physics and graph theory to investigate financial markets.
Dr Di Matteo came from the University of Salerno in Italy to ANU to establish a research centre in this intriguing — and sophisticated — area of physics, well-established in parts of Europe and the United States, but less so in the southern hemisphere.
Based in the Department of Applied Mathematics in the Research School of Physical Sciences and Engineering, Dr Di Matteo says the complex blend of economics and physics sates her aspiration for challenging and rewarding research.
“I started my career as a condensed matter physicist, investigating the properties of superconducting materials,” Dr Di Matteo says.
“But little by little my interest moved toward the science of complex systems. I realised that very often natural systems are intrinsically complex and the disentanglement of such a complexity is probably the most challenging, and potentially rewarding, scientific topic of the present era.”
As much as Dr Di Matteo needs to understand the physics of complex systems, she is also well read on financial literature — concepts like portfolio optimisation, Federal funds rate fluctuations and risk hedging are commonly handled by Econophysicists — the study of econophysics is not as esoteric as it might first appear.
Wealth
“A better understanding of the properties of financial markets is of great importance: the lives of most of us depend on the dynamics of financial markets that affect investments, savings, business, employment, growth, wealth and — ultimately — the daily functioning of our society,” Dr Di Matteo says.
“The goal is to understand the hidden mechanisms that lead to emergence of patterns in the markets. For instance, we have shown that the multifractal properties of price fluctuations are deeply related to the stage of development of the markets, a fact that can be used to differentiate between markets and help investors to hedge their risk.
“But from a purely scientific point of view — and the reason why I enjoy this field — is that financial markets are perfect examples of complex systems where a large number of agents interact in a complicated way, making actions which span over several orders of magnitude in size and time.
“To me, the surprising and fascinating aspect is that in such intricate systems we observe the emergence of behaviours and patterns that appear to be universal and common to several other complex systems.”
 |
A 3-D representation of 34 different interest rates from the US Federal Reserve database. Again each node (circle) represents an interest rate and each edge-length corresponds to the correlation between two interest rates. |
In between research commitments, Dr Di Matteo and fellow econophysicist Dr Tomaso Aste are busy organising Australia’s own international econophysics conference, which will bring together experts in this field from around the world, including founders of econophysics, Professor Eugene Stanley from Boston University and Professor Rosario Mantegna from Palermo University.
“This conference will be one of the keys to establishing an Australian presence in the field of econophysics — there is a lot of talent in the physical sciences in this part of the world, so we hope to attract some of this talent into this emerging field in Australia,” Dr Di Matteo says.
Productivity
In recent research Dr Di Matteo has shown how that the way financial firms are connected to each other influences productivity and uptake of technology.
The more ”network connected” the firms are, the more efficiently information disseminates between the firms and the faster productivity growth will be.
Network connection is a key element in the study of complex systems – these are the links between the interacting elements that make an organised system. As a result, various financial markets actually show similar patterns, behaviours and trends despite the influence of thousands of random variables. These patterns and trends can be identified, classified and eventually exploited.
In a recent paper, Dr Di Matteo and others analysed interest rate data and demonstrated how distinct financial products behave collectively in a hierarchical structure.
By conducting innovative analysis of the structure of correlations between interest rates of each financial product, they found that the differentiation between specific properties emerges automatically.
The organisation of these structures has delivered some fascinating models of financial market behaviour that have been rendered into three-dimensional images by the ANU Supercomputer Facility’s Vizlab.
“Generally, the properties of complex systems are determined by the structure of interactions between many elements,” Dr Di Matteo says.
“The behaviour of each element is meaningless: what is relevant is the collective dynamics of the system as a whole. The 3-D images give a quantitative representation of such a collective dynamics and have been designed to catch all the relevant properties in a simple and intuitive way.”
More:
www.rsphysse.anu.edu.au/econophysics
|
