Dependency mining in large sets of stock market trading rules

Authors:
Piotr Lipinski
Affiliations:
Institute of Computer Science, University of Wroclaw, Wroclaw, Poland and LSIIT, CNRS, Université Louis Pasteur, Strasbourg, France
Venue:
Enhanced methods in computer security, biometric and artificial intelligence systems
Year:
2005

Citing 2
Cited 4

Self-organizing maps

Self-organizing maps
Evolution Strategy in Portfolio Optimization

Selected Papers from the 5th European Conference on Artificial Evolution

Evolutionary Decision Support System for Stock Market Trading

AIMSA '08 Proceedings of the 13th international conference on Artificial Intelligence: Methodology, Systems, and Applications
Evolution strategies for objective functions with locally correlated variables

IDEAL'10 Proceedings of the 11th international conference on Intelligent data engineering and automated learning
Evolutionary approach to multiobjective optimization of portfolios that reflect the behaviour of investment funds

AIMSA'12 Proceedings of the 15th international conference on Artificial Intelligence: methodology, systems, and applications
Usage patterns of trading rules in stock market trading strategies optimized with evolutionary methods

EvoApplications'13 Proceedings of the 16th European conference on Applications of Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses the problem of dependency mining in large sets. The first goal is to determine and reduce the dimension of data using principal component analysis. The second is to group variables into several classes using Kohonen's self-organizing maps and then the K-means algorithm. Evaluations have been performed on 350 financial trading rules (variables) observed in a period of 1300 instants (observations). It was shown that the rules are strongly correlated, all of which can be reproduced from 150 generators with an accuracy of 95%. Moreover, the initial set of 350 rules was subdivided into 23 classes of similar rules.