Title
Forex Trading Using Geometry Sensitive Neural Networks
Keywords
DXNN; Evolutionary computation; Financial analysis; Forex; Memetic algorithm; Neural network; Neuroevolution
Abstract
When neural network based systems are used within the field of financial analysis, either as price oracles or autonomous traders, they are primarily used with a sliding price window. This paper presents a novel approach where the indirectly encoded neural network system, just like the technical analysts, looks directly at the candlestick style sliding chart instead, the actual geometrical-patterns within it, to make its predictions. The results presented demonstrate that this approach results in a higher and more consistent generalization to previously unseen financial data, while maintaining a profit level on par with the neuroevolutionary system which uses a standard sliding window. Copyright is held by the author/owner(s).
Publication Date
1-1-2012
Publication Title
GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion
Number of Pages
1533-1534
Document Type
Article; Proceedings Paper
Personal Identifier
scopus
DOI Link
https://doi.org/10.1145/2330784.2331032
Copyright Status
Unknown
Socpus ID
84864974247 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/84864974247
STARS Citation
Sher, Gene I., "Forex Trading Using Geometry Sensitive Neural Networks" (2012). Scopus Export 2010-2014. 5675.
https://stars.library.ucf.edu/scopus2010/5675