High-frequency Finance Project

The most promising approach to study financial markets is through observation and modelling micro-changes in the market.

People | Expertise | Directional Change Project | Demos | HFF Platform || Affiliations: CCFEA | BRACIL || Sponsor:   Olsen Ltd / Lykke

The Global Challenge

The current economic crisis is a testament of the deficiency of modern economic theory and its tools of analysis. In response to the current economic crisis, Gordon Brown and Angela Merkel among others have called for the development of a global early warning system. Our proposal is a response to this call of action. To our knowledge, there does not exist a global information system that can fulfil the requirements of system proposed by the politicians. The proposed approach leverages recent discoveries that a group of practitioners have started to utilize with success. Financial markets affect the economy, politics, and most people’s livelihood. Yet, we know very little about the behaviour of the markets. For example, under what conditions do crashes happen? Can we build an early warning system against crashes? How can the market mechanism be improved to avoid excessive volatility? Theories on volatility have been established, but we know that they do not fully explain the observed market behaviour. Are there alternative ways forward?

Storms rarely happen without signs. The question is whether one can pick up those signs. In weather forecasting, data about micro changes are collected by sensors, which are used to form bigger pictures and make predictions. Forecasts are never perfect, as our models on changes are not. Nevertheless, a micro-changes approach represents our best shot at weather forecasting today. It is rather surprising to see that this approach is relatively untouched in economics.

Price movements of financial markets are the result of their structure, the behaviour of market participants, the balance of supply and demand and pending limit order. We do not know a priori, when traders make buy or sell orders and cannot anticipate the size of the next price move. There is, however, a lot of information that is available that has not been researched. All trading activities from opening to closing of positions are recorded, tick-by-tick. In fact, all orders, including those that are not executed, are recorded. With this information, it is possible to build models of trader behaviour and gain an insight, how in aggregate positions in the market built up and closed out again. The process has an impact on the price evolution and is a valuable tool to make inferences about the future price evolution. Detailed studies of trading behaviour and price movements give us a better understanding of market behaviour. For example, what is the impact of big orders, how much over-shooting do they cause? Large movements force leveraged traders to close out their positions, which brings about a cascade of further margin calls accelerating the price movements There is a wealth of data sitting there that needs to be studied. Very little has been done so far.

Our proposal

We plan to build a research platform for a global information system of the economy and the financial markets. The platform utilizes the discoveries of high frequency finance to develop an innovative approach of economic modelling and analysis of trader behaviour by combing a broad range of disciplines. (read more...)

Maintained by: Edward Tsang; Created: 2009.09.16 Last Updated: 2016.04.04