Research Article
JPX Tokyo Stock Exchange Prediction with Deep Neural Networks
@INPROCEEDINGS{10.4108/eai.12-1-2024.2347128, author={Shurui Hu and Hua Hansen and Junchao Mao and Jiabing Zhang}, title={JPX Tokyo Stock Exchange Prediction with Deep Neural Networks}, proceedings={Proceedings of the 3rd International Conference on Big Data Economy and Digital Management, BDEDM 2024, January 12--14, 2024, Ningbo, China}, publisher={EAI}, proceedings_a={BDEDM}, year={2024}, month={6}, keywords={jpx tokyo stock exchange quantitative investment dnn sharpe ratio}, doi={10.4108/eai.12-1-2024.2347128} }- Shurui Hu
Hua Hansen
Junchao Mao
Jiabing Zhang
Year: 2024
JPX Tokyo Stock Exchange Prediction with Deep Neural Networks
BDEDM
EAI
DOI: 10.4108/eai.12-1-2024.2347128
Abstract
Quantitative investment is a quantitative analysis method, according to the historical related data, the use of computer technology combined with mathematical models to establish a quantitative analysis model of investment, to discover the price: the hidden trend behind the change, so as to guide the formulation of investment strategies, so it has been widely used in the field of investment. In this paper, we focus on the JPX stock exchange prediction using deep learning algorithms. We adopt the Deep Neural Network (DNN) as our model, and we do compared experiments with some classical models. Our DNN model owns the best performance in teams of the highest Sharpe Ratio score 0.158. On the contrary, The Xgboost and Lightgbm respectively owns 0.107, 0.112 Sharpe Ratio which are all lower than DNN’s Sharpe Ratio.
