Fast training of support vector machines using sequential minimal optimization
Advances in kernel methods
Machine Learning
An Empirical Analysis of Data Requirements for Financial Forecasting with Neural Networks
Journal of Management Information Systems
Data stream mining for market-neutral algorithmic trading
Proceedings of the 2008 ACM symposium on Applied computing
Autonomous Forex Trading Agents
ICDM '08 Proceedings of the 8th industrial conference on Advances in Data Mining: Medical Applications, E-Commerce, Marketing, and Theoretical Aspects
Foreign-Exchange-Rate Forecasting with Artificial Neural Networks
Foreign-Exchange-Rate Forecasting with Artificial Neural Networks
Optimization of the trading rule in foreign exchange using genetic algorithm
Proceedings of the 11th Annual conference on Genetic and evolutionary computation
The WEKA data mining software: an update
ACM SIGKDD Explorations Newsletter
An Automated Trading System with Multi-indicator Fusion Based on D-S Evidence Theory in Forex Market
FSKD '09 Proceedings of the 2009 Sixth International Conference on Fuzzy Systems and Knowledge Discovery - Volume 03
Evaluating the performance of adapting trading strategies with different memory lengths
IDEAL'09 Proceedings of the 10th international conference on Intelligent data engineering and automated learning
Estimating continuous distributions in Bayesian classifiers
UAI'95 Proceedings of the Eleventh conference on Uncertainty in artificial intelligence
Financial volatility trading using recurrent neural networks
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Automated trading systems for financial markets can use data mining techniques for future price movement prediction. However, classifier accuracy is only one important component in such a system: the other is a decision procedure utilizing the prediction in order to be long, short or out of the market. In this paper, we investigate the use of technical indicators as a means of deciding when to trade in the direction of a classifier's prediction. We compare this "hybrid" technical/data stream mining-based system with a naive system that always trades in the direction of predicted price movement. We are able to show via evaluations across five financial market datasets that our novel hybrid technique frequently outperforms the naive system. To strengthen our conclusions, we also include in our evaluation several "simple" trading strategies without any data mining component that provide a much stronger baseline for comparison than traditional buy-and-hold or sell-and-hold strategies.