NECSI Resources

 International Conference on Complex Systems (ICCS2007)

Computational Intelligence in Financial Contagion Analysis

Antoaneta Serguieva
Business School of Brunel University

Hao Wu
Business School of Brunel University

     Full text: PDF
     Last modified: December 18, 2007

Abstract

Problem: Over the past two decades, financial market crises have occurred in different regions and revealed a unifying characteristic. Whether the crisis is in Mexico, Asia, or Russia, the financial difficulties from one economy spread to neighbouring economies in the region. If the difficulties are spread through stable cross-market linkages, then the country experiencing the original shock can take measures improving the fundamentals. If the shocks are propagated although the fundamentals are sound, then an IMF intervention will be appropriate. Unstable cross-market linkages during financial crises are referred to as financial contagion. Financial contagion has been studied intensively in recent years, focusing on the definition and the statistical measuring of the phenomenon, and the ex-post investigation of crisis data for the existence of contagion. Alternatively, we consider here simulating financial markets and linkages through modelling market participants and their various strategies. Modifying the agents’ behaviours and evolving linked markets will allow testing for financial contagion under various scenarios.

Methods: We develop an agent-based model and investigate herding behaviours as a reason for contagion. Participants in financial markets are divided as smart traders and noise traders. Smart traders make decisions based on fundamental or technical analysis, while noise traders contribute to herding behaviours. The starting point is a linked-market model based on statistical mechanics, with varying ratios of smart vs. noise traders in each market. To be able to study more realistic strategies, we develop the model further implementing a minority game approach.

Results: Various strategies and simulation parameters are able to divide agents into different types. We also explore different parameters and market characteristics. Through introducing shocks to one of the markets, the investigation is focused on the simulations which present evidence for financial contagion.

Conclusions: Though a problem of current interest, financial contagion has not been studied using computational intelligence techniques. It is a phenomenon occurring through the function and linkages of financial markets as complex systems, and should be approached using the range of methodologies available within the complex systems paradigm. Investigating the characteristics of the simulations leading to financial contagion, will contribute to recognising at an earlier stage financial crises which potentially destabilise cross-market linkages. In the real world, such information would be valuable in enabling appropriate risk management actions. Our future research will focus on increasing the complexity of the system through augmenting the minority game with evolutionary computation.

Bibliography
[1] Alfarano, S., Lux, T., & Wagner, F., 2006, Estimation of Simple Agent-Based Model of Financial Market: An Application to Australian Stock and Foreign Exchange Data, Physica A, Vol. 370(1), pp 38-42.
[2] Cont, R., & Bouchaud, J.P., 2000, Herd Behavior and Aggregate Fluctuations in Financial Market, Macroeconomic Dynamics, Vol. 4, pp 170-196.
[3] Corsetti, G., Pericoli, M., & Sbracia, M., 2005, ‘Some Contagion, Some Interdependence’: More Pitfalls in Tests of Financial Contagion, Journal of International Money and Finance, Vol. 24, pp 1177-1199.
[4] Caporale, G.M., Cipollini, A., & Spagnolo, N., 2005, Testing for Contagion: A Conditional Correlation Analysis, Journal of Empirical Finance, Vol. 12, pp 476-489.
[5] Caporale, G.M., Pittis, N., & Spagnolo, N., 2006, Volatility Transmission and Financial Crises, Journal of Economics and Finance, Vol. 30, pp 376-390.
[6] Challet, D., & Zhang, Y-C., 1997, Emergence of Cooperation and Organization in an Evolutionary Game, Physica A, Vol. 256, pp 514-532.
[7] Challet, D., Marsilli, M., & Zhang, Y-C., 2000, Modeling Market Mechanism with Minority Game, Physica A, Vol. 276(1), pp 284-315.
[8] Challet, D., Marsilt, M., & Zhang Y-C., 2005, Minority Games, Oxford University Press (Oxford).
[9] De Long, J.B., Shleifer, A., Summers, L.H., & Waldmann, R.J., 1999, Noise Trader Risk in Financial Market, Journal of Political Economy, Vol. 98(4), pp 703-738.
[10] Fogel, D.B., & Stayton, L.C., 1994, On the Effectiveness of Crossover in Simulated Evolutionary Optimization, Biosystems, Vol. 2(3), pp 171-182.
[11] Forbes, K., & Rigobon, R., 2002, No Contagion, Only Interdependence: Measuring Stock Market Comovements, The Journal of Finance, Vol. 57, pp 2223-2261.
[12] Kaizoji, T., 2001, A Model of International Financial Crises, Physica A,
Vol. 299, pp 279-293.
[13] LeBaron, B., 2000, Agent Based Computational Finance: Suggested Readings and Early Research, Journal of Economic Dynamics and Control, Vol. 24, pp 679-702.
[14] Tanaka-Yamawaki, M., & Tokuoka, S., 2006, Minority Game as a Model for the Artificial Financial Markets, 2006 IEEE Congress on Evolutionary Computation in Vancouver, July 16-21, pp 2157-2162
[15] Sysi-Aho, M., Chakraborti, A., & Kaski, K., 2003, Adaptation Using Hybridized Genetic Crossover Strategies, Physica A, Vol. 322, pp 701-709.








Maintained by NECSI Webmaster Copyright © 2000-2007 New England Complex Systems Institute. All rights reserved.