Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorLim, Wee
dc.description.abstractFor the purpose of this research, three machine learning strategies for trading were studied and implemented in order to effectively come up with a conclusion for the Foreign Exchange Market (Forex) of EUR/USD currency pair regarding which strategy is most profitable. The strategies used in this project were ANFIS, SVM and KNN. The constructed models then attempted to imitate the behaviors, responses as well as decision making skills of a human in the Forex market. To accurately portray the human decision, a set of technical indicators which are commonly used by traders were analyzed to represent it in this experiment. The models were all trained and tested with sets of data processed based on this set of technical indicators. The models were then evaluated based on their misclassification error and total profit-or-loss based on a unit of open price. Upon analyzing the output produced, it is possible to conclude that among the three strategies, using data from the technical indicators to train the models, SVM would be the most profitable compared to ANFIS and KNN in terms of making the correct decision in the Forex market to maximize profits and minimize losses.en_US
dc.format.extent100 p.en_US
dc.rightsNanyang Technological University
dc.subjectDRNTU::Engineering::Computer science and engineeringen_US
dc.titleMachine learning strategies for tradingen_US
dc.typeFinal Year Project (FYP)en_US
dc.contributor.supervisorRavi Suppiahen_US
dc.contributor.schoolSchool of Computer Engineeringen_US
dc.description.degreeBachelor of Engineering (Computer Science)en_US
item.fulltextWith Fulltext-
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)
Files in This Item:
File Description SizeFormat 
  Restricted Access
SCE15-0424 - Machine Learning Strategies for Trading2.83 MBAdobe PDFView/Open

Page view(s)

Updated on Jul 20, 2024

Download(s) 50

Updated on Jul 20, 2024

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.