Virtual explanation for market cycles

QUEEN Elizabeth recently asked economists “why didn’t anybody see this coming” in relation to the global downturn, and now some…

QUEEN Elizabeth recently asked economists “why didn’t anybody see this coming” in relation to the global downturn, and now some Galway researchers are trying to answer that question.

A “virtual stock market” has been initiated by NUI Galway (NUIG) academics to try to understand the “boom and bust” cycles of financial markets.

The virtual market has employed human traders and automated computer agents for the series of experiments, conducted by NUIG’s JE Cairnes School of Business and Economics, the Digital Enterprise Research Institute (DERI) and the Computer Integrated Manufacturing Research Unit (CIMRU).

The market lists 10 companies and has four types of computer agents trading. Participating NUIG students have been given virtual currency to the value of 10,000 “airgead” and equivalent shares.

READ MORE

NUIG economics lecturer Dr Srinivas Raghavendra says the objective is to “understand the generating processes that underlie the empirical facts of the real-world financial markets”.

The traders are invited to try their investment strategies, technical trading strategies or other hybrid strategies. The real time data of the virtual market can be directly downloaded to a spreadsheet, that will allow participants to try out various technical trading rules, the research team says.

The trading data is available to the “traders” at two frequencies. The team aims to test the experiments in a professional stock exchange environment and the virtual market will also be applied as a teaching tool for students.

“Even the anticipation of a monetary policy swings the markets, let alone the post announcement effect,” Dr Raghavendra says.

“The experiments in this area would be useful to study the dynamic between monetary policy rules and instruments, and human agents’ expectation formation, which is one of the fundamental issues in understanding the stability of markets,” he says.

Results of the trials are due to be presented at the 15th Inter-national Conference on Computing in Economics and Finance in Sydney, Australia in July.