TITLE: Trading and Market Surveillance in the Capital Markets: Can machine learning help?

PRESENTER: Richard Coggins

AFFILIATION: School of Electrical and Information Engineering, University of Sydney and the Capital Markets CRC

DATE: Friday 18th June 2004

TIME: 12:00noon - 1.00pm

PLACE: CSE K17 1st Floor Seminar Room

ABSTRACT:

Since the late 1990's, as most securities markets converted to fully electronic operation, extensive financial transaction data bases have become available for research. The Capital Markets CRC with the support of it's industry partners is leveraging such a data repository to research and develop optimised trading strategies for exchange traded securities and automated classification systems to identify anomalous trading behaviour for security market surveillance. The databases we are working with include ASX SEATS comprising full intra day order book data and Reuters data for over 200 world markets at the trade and quote level. These transaction databases are augmented by announcement databases such as ASX Signal G to explore the relationship between announcements and market reaction.

Our research on intra-day trading strategies will be presented by means of a demonstration of a trading evaluation environment. I'll outline some of the key research questions for trading and surveillance in security markets. These include identifying anomolous trading behaviours such as front running, market manipulation and insider trading, optimising large institutional trade execution strategies to minimise transaction costs, and estimating market responses to other traders actions and corporate information releases. While introducing these research areas I'll indicate potential applications of machine learning techniques to these problems.

BIOGRAPHY OF SPEAKER:

Richard Coggins is currently a senior lecturer in computer engineering in the School of Electrical and Information Engineering at the University of Sydney. He holds a PhD from the University of Sydney addressing the areas of pattern recognition and time series analysis applied to biomedical time series, a bachelor of engineering honours I. (electrical) and a bachelor of science (physics/pure mathematics) and is a member of the Institute of Electrical and Electronics Engineers, USA. Since July 2001, he has been a researcher in the Data Mining program of the Capital Markets CRC, with projects addressing trading strategies, trading analytics and market microstructure. His current research interests include adaptive agents, machine learning, time series analysis and pattern recognition with a focus on applying these technologies to the capital markets for trading and surveillance.world.

Host & Seminar Convenor:

Bernhard Hengst
Tel: +61-2-9385-3988,
E-mail: bernhardh@cse.unsw.edu.au
School of Computer Science & Engineering, UNSW.