Identification of factors characterising volatility and firm-specific risk using ensemble classifiers

Authors:
Pascal Khoury;Denise Gorse
Affiliations:
Dept. of Computer Science, University College London, London, UK,Charlemagne Capital, London, UK;Dept. of Computer Science, University College London, London, UK
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part IV
Year:
2012

Citing 2
Cited 0

A review of particle swarm optimization. Part II: hybridisation, combinatorial, multicriteria and constrained optimization, and indicative applications

Natural Computing: an international journal
A comparison of GA and PSO for excess return evaluation in stock markets

IWINAC'05 Proceedings of the First international work-conference on the Interplay Between Natural and Artificial Computation conference on Artificial Intelligence and Knowledge Engineering Applications: a bioinspired approach - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Ensemble classifiers comprised of neural networks trained using particle swarm optimisation are used to identify characteristic factors of companies that have experienced high share price volatility either by an objective measure or relative to their peers, or whose calculated firm-specific equity risk exceeds a comparison value appropriate to their industry and region. Use is made of a novel training metric, the Matthews correlation coefficient, that is shown to better handle numerically unbalanced data sets. A comparison is made with results from input-output correlation analysis and it is noted that the factors derived from ensemble weightings appear to be more predictive.