Research Article

NEURO-GENETIC PREDICTIONS OF CURRENCY CRISES

  1. Peter Sarlin1,*,
  2. Dorina Marghescu2

Article first published online: 13 NOV 2011

DOI: 10.1002/isaf.328

Issue

Intelligent Systems in Accounting, Finance and Management

Intelligent Systems in Accounting, Finance and Management

Volume 18, Issue 4, pages 145–160, October/December 2011

Additional Information

How to Cite

Sarlin, P. and Marghescu, D. (2011), NEURO-GENETIC PREDICTIONS OF CURRENCY CRISES. Int. J. Intell. Syst. Acc. Fin. Mgmt., 18: 145–160. doi: 10.1002/isaf.328

Author Information

  1. 1

    Turku Centre for Computer Science – TUCS, Department of Information Technologies, Åbo Akademi University, Turku, Finland

  2. 2

    Centre for Knowledge and Innovation Research (CKIR), Aalto University School of Economics, Helsinki, Finland

*Peter Sarlin, Turku Centre for Computer Science – TUCS, Department of Information Technologies, Åbo Akademi University, Turku, Finland. E-mail: psarlin@abo.fi

Publication History

  1. Issue published online: 21 FEB 2012
  2. Article first published online: 13 NOV 2011

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

View Full Article (HTML) Get PDF (283K)

Keywords:

  • neuro-genetic model;
  • currency crisis;
  • prediction;
  • artificial neural networks;
  • genetic algorithms

SUMMARY

We create a neuro-genetic (NG) model for predicting currency crises by using a genetic algorithm for specifying (1) the combination of inputs, (2) the network configuration and (3) the training parameters for a back-propagation artificial neural network (ANN). The performance of the NG model is evaluated by comparing it with standalone probit and ANN models in terms of utility for a policy decision-maker. We show that the NG model provides better in-sample and out-of-sample performance, as well as provides an automatic and more objective calibration of a predictive ANN model. We show that using a genetic algorithm for finding an optimal model specification for an ANN is not only less laborious for the analyst, but also more accurate in terms of classifying in-sample and predicting out-of-sample crises. For a sufficiently parsimonious, but still nonlinear, model for generalized processing of out-of-sample data, the creation and evaluation of models is performed objectively using only in-sample information as well as an early stopping procedure. Copyright © 2011 John Wiley & Sons, Ltd.

View Full Article (HTML) Get PDF (283K)