Forex trend classification using machine learning techniques

Authors:
Areej Abdullah Baasher;Mohamed Waleed Fakhr
Affiliations:
Computer Science Department, Arab Academy for Science and Technology, Cairo, Egypt;Computer Science Department, Arab Academy for Science and Technology, Cairo, Egypt
Venue:
ACS'11 Proceedings of the 11th WSEAS international conference on Applied computer science
Year:
2011

Citing 5
Cited 0

Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Trading on the Edge: Neural, Genetic, and Fuzzy Systems for Chaotic Financial Markets

Trading on the Edge: Neural, Genetic, and Fuzzy Systems for Chaotic Financial Markets
An introduction to variable and feature selection

The Journal of Machine Learning Research
Variable selection using svm based criteria

The Journal of Machine Learning Research
Quarterly Time-Series Forecasting With Neural Networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Foreign Currency Exchange market (Forex) is a highly volatile complex time series for which predicting the daily trend is a challenging problem. In this paper, we investigate the prediction of the High exchange rate daily trend as a binary classification problem, with uptrend and downtrend outcomes. A large number of basic features driven from the time series data, including technical analysis features are generated using multiple history time windows. Various feature selection and feature extraction techniques are used to find best subsets for the classification problem. Machine learning systems are tested for each feature subset and results are analyzed. Four important Forex currency pairs are investigated and the results show consistent success in the daily prediction and in the expected profit.