Foreign exchange prediction using the long short-term memory neural network

• المؤلف الرئيسي: Zheng, Xiaojun
• مؤلفون آخرون: Wang Lipo
• التنسيق: Final Year Project
• اللغة: English
• منشور في: 2019
• الموضوعات: DRNTU::Engineering::Electrical and electronic engineering
• الوصول للمادة أونلاين: http://hdl.handle.net/10356/77377
• المؤسسة: Nanyang Technological University

The foreign exchange (Forex)is closely related to our life, for example when we travel abroad, we need the currency of the destination country, for currency traders, they can even earn the profit on currency spreads. The foreign exchange market is very active; many factors will affect the foreign exchange rate, for example, Inflation, Government Debt, Political Stability & Performance and so on [1]. With the rapid development of technology, artificial neural network (ANN) technology has been widely used in various fields; there are many kinds of ANN, such as Multilayer Perceptrons (MLP), Convolutional Neural Network (CNN) and Recurrent Neural Networks (RNN). This project goal is to explore foreign exchange prediction and trading by using the long short-term memory neural network (LSTM), showing that the accuracy and effectiveness of the proposed method. This project will be using numpy, pandas, Tensorflow, Keras and Matplotlib. By implemented those functions to achieve the goal. The input of the LSTM model will be the closing price of the USD/JPY, AUD/JPY, EUR/USD, and GBP/USD.