Open AccessApplication of LSTM Neural Network in Forecasting Foreign Exchange Price
Author(s) -
Yingming Qu,
Xue Fang Zhao
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1237/4/042036
Subject(s) - mean squared error , artificial neural network , recurrent neural network , foreign exchange , computer science , artificial intelligence , foreign exchange market , deep learning , time series , series (stratigraphy) , mean absolute error , mean squared prediction error , machine learning , mathematics , statistics , economics , monetary economics , paleontology , biology
LSTM neural network and RNN neural network models in deep learning are used to forecast the price of foreign exchange financial time series. The existing foreign exchange price and technical analysis indexes are taken as input parameters. By comparing the evaluation indexes of two deep learning models, the optimal neural network model is selected. The experimental results show that the LSTM neural network model has smaller root mean square error (RMSE) and mean absolute error (MAE) than the RNN network model, and the predicted price is more accurate.