| Summary: | 碩士 === 國立東華大學 === 資訊工程學系 === 106 === FX market has tens of thousands of transactions every day. This paper hopes to find regularities in the FX market via machine learning. This paper uses deep learning Convolutional Neural Network ( CNN ) and Long Short-Term Memory ( LSTM ) models for FX analysis.
This paper establishes the CNN and the LSTM model to forecast, and the two cooperate with each other and set the trading conditions. The experimental use of the sample period USD/JPY from January 10, 2005 to March 30, 2018, and from January 2, 2018 to March 30, 2018, the results which can earn profit in the FX.
|
|---|
