Advancing the Backtrack Optimization Technique to Obtain Forecasts of Potential Crisis Periods

Abstract

Financial crisis is an unfortunate reality that overshadows any financial system regardless its profitability and the level it functions. The appearance of crises across financial markets, especially during the 1990s that the internationalized markets adopted a rather approachable character, imposed severe costs in financial and social systems. With this paper is proposed the generation of a future interval of time that is vulnerable to enclose the burst of a financial crisis. A time series consisted of approximations of the local Lipschitz constant is examined and in the proposed forecasting approach this constant holds the crisis indicator role. Further the application of two different optimization techniques over the Lipschitz-made time series results to the generation of a future period of time; this interval is likely to envelop the burst of a forthcoming crisis. The usage of a future interval of time empowers the predicting ability of the methodology by providing warning signs priory to the actual crisis burst. To this direction, the obtained results offer strong evidence that the method may be characterized as an Early Warning System (EWS) for financial crisis prediction.

Financial crisis is an unfortunate reality that overshadows any financial system regardless its profitability and the level it functions. The appearance of crises across financial markets, especially during the 1990s that the internationalized markets adopted a rather approachable character, imposed severe costs in financial and social systems. With this paper is proposed the generation of a future interval of time that is vulnerable to enclose the burst of a financial crisis. A time series consisted of approximations of the local Lipschitz constant is examined and in the proposed forecasting approach this constant holds the crisis indicator role. Further the application of two different optimization techniques over the Lipschitz-made time series results to the generation of a future period of time; this interval is likely to envelop the burst of a forthcoming crisis. The usage of a future interval of time empowers the predicting ability of the methodology by providing warning signs priory to the actual crisis burst. To this direction, the obtained results offer strong evidence that the method may be characterized as an Early Warning System (EWS) for financial crisis prediction.

Cite this paper

E. Lisgara, G. Karolidis and G. Androulakis, "Advancing the Backtrack Optimization Technique to Obtain Forecasts of Potential Crisis Periods,"*Applied Mathematics*, Vol. 3 No. 10, 2012, pp. 1538-1551. doi: 10.4236/am.2012.330214.

E. Lisgara, G. Karolidis and G. Androulakis, "Advancing the Backtrack Optimization Technique to Obtain Forecasts of Potential Crisis Periods,"

References

[1] G. Kaminsky, S. Lizondo and C. M. Reinhart, “Leading Indicators of Currency Crises,” International Monetary Fund Staff Papers, Vol. 1, No. 1, 1998, pp. 1-48.
doi:10.2307/3867328

[2] A. Berg and C. Pattillo, “Predicting Currency Crises: The Indicators Approach and an Alternative,” Journal of International Money and Finance, Vol. 18, No. 4, 1999, pp. 561-586. doi:10.1016/S0261-5606(99)00024-8

[3] E. G. Lisgara and G. S. Androulakis, “Estimating Time Series Future Optima Using a Steepest Descent Methodology as a Backtracker,” Proceedings of the International Multiconference on Computer Science and Information Technology, Wisla, 20-22 October 2008, pp. 893-898.

[4] G. S. Androulakis and E. G. Lisgara, “Exploiting the Usage of the Lipschitz Constant Approximations as a Financial Crisis Indicator,” submitted, 2011.

[5] E. G. Lisgara, G. I. Karolidis, and G. S. Androulakis, “Entrapping a Time Series Future Optima Using a Combination of Optimization Techniques,” Proceedings of the 24 Mini EURO Conference on Continuous Optimization and Information-Based Technologies in The Financial Sector, Izmir, 23-26 June 2010, pp. 24-29.

[6] G. Kaminsky and C. M. Reinhart, “The Twin Crises: The causes of Banking and Balance-of-Payments Problems,” American Economic Review, Vol. 89, No. 3, 1999, pp. 473-500. doi:10.1257/aer.89.3.473

[7] R. Majhi, G. Panda and G. Sahoo, “Efficient prediction of Exchange Rates with Low Complexity Artificial Neural Network Models,” Expert Systems with Applications, Vol. 36, No. 1, 2009, pp. 181-189.
doi:10.1016/j.eswa.2007.09.005

[8] R. Dornbusch, I. Goldfajn and R. O. Valdes, “Currency Crises and Collapses,” Brookings Papers on Economic Activity, Vol. 26, No. 2, 1995, pp. 219-294.
doi:10.2307/2534613

[9] P. Krugman, “Are Currency Crises Self-Fulfilling?” NBER Macroeconomics Annual, Vol. 11, 1996, pp. 345-407.
doi:10.2307/3585207

[10] V. Coudert and M. Gex, “Does Risk Aversion Drive Financial Crises? Testing the Predictive Power of Empirical Indicators,” Journal of Empirical Finance, Vol. 15, No. 2, 2007, pp. 167-184. doi:10.1016/j.jempfin.2007.06.001

[11] M. Bussiere and M. Fratzscher, “Towards a New Early Warning System of Financial Crises,” Journal of International Money and Finance, Vol. 25, No. 6, 2006, pp. 953-973. doi:10.1016/j.jimonfin.2006.07.007

[12] A. Berg and R. E. Coke, “Autocorrelation-Corrected Standard Errors in Panel Probits: An Application to Cur-Rency Crisis Prediction,” Technical Report: International Monetary Fund, 2004.

[13] J. Berg, B. Candelon and J. Urbain, “A Cautious Note on the Use of Panel Models to Predict Financial Crises,” Economics Letters, Vol. 101, No. 2, 2008, pp. 80-83.
doi:10.1016/j.econlet.2008.06.015

[14] M. Abramowitz and I. A. Stegun, “Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables,” National Bureau of Standards Applied Mathematics, 1997.

[15] A. Molinaro, C. Pizzuti and Y. D. Sergeyev, “Acceleration Tools for Diagonal Information Global Optimization Algorithms,” Computational Optimization and Applications, Vol. 18, No. 1, 2001, pp. 5-26.
doi:10.1023/A:1008719926680

[16] E. G. Lisgara, G. I. Karolidis and G. S. Androulakis, “A Progression of the Backtrack Optimization Technique Forforecasting Potential Financial Crisisperiods,” Proceedings of the Conference in Numerical Analysis, Recent Approaches to Numerical Analysis: Theory, Methods and Applications, 15-18 September 2010, pp. 126-133.

[17] G. S. Androulakis and M. N. Vrahatis, “OPTAC: A Portable Software Package for Analyzing and Comparing Optimization Methods by Visualization,” Journal of Computational and Applied Mathematics, Vol. 72, No. 1, 1996, pp. 41-62. doi:10.1016/0377-0427(95)00244-8

[18] G. S. Androulakis, “A Technique for Entrapping a Time Series Future Optima,” Proceedings of the Conference in Numerical Analysis, Recent Approaches to Numerical Analysis: Theory, Methods and Applications, 15-18 September 2010, pp. 19-23.

[19] G. I. Karolidis, E. G. Lisgara and G. S. Androulakis, “A New Projection Method for Predicting Time Series Points,” Proceedings of the 24 Mini EURO Conference on Continuous Optimization and Information Based Technologies in The Financial Sector, Izmir, 23-26 June 2010, pp. 18-23.

[20] G. S. Androulakisand and E. G. Lisgara, “On the Prediction of Time Series’ Local Optima: A Backtrack Technique,” Proceedings of the Conference in Numerical Analysis, Recent Approaches to Numerical Analysis: Theory, Methods and Applications, Kalamata, 3-7 September 2007, pp. 19-23.

[21] R Development Core Team, “R: A Language and Environment for Statistical Computing,” R Foundation for Statistical Computing, Vienna, 2008.
http://www.R-project.org